1 // SPDX-License-Identifier: GPL-2.0-or-later 2 /* 3 * SR-IPv6 implementation 4 * 5 * Authors: 6 * David Lebrun <david.lebrun@uclouvain.be> 7 * eBPF support: Mathieu Xhonneux <m.xhonneux@gmail.com> 8 */ 9 10 #include <linux/types.h> 11 #include <linux/skbuff.h> 12 #include <linux/net.h> 13 #include <linux/module.h> 14 #include <net/ip.h> 15 #include <net/lwtunnel.h> 16 #include <net/netevent.h> 17 #include <net/netns/generic.h> 18 #include <net/ip6_fib.h> 19 #include <net/route.h> 20 #include <net/seg6.h> 21 #include <linux/seg6.h> 22 #include <linux/seg6_local.h> 23 #include <net/addrconf.h> 24 #include <net/ip6_route.h> 25 #include <net/dst_cache.h> 26 #include <net/ip_tunnels.h> 27 #ifdef CONFIG_IPV6_SEG6_HMAC 28 #include <net/seg6_hmac.h> 29 #endif 30 #include <net/seg6_local.h> 31 #include <linux/etherdevice.h> 32 #include <linux/bpf.h> 33 34 #define SEG6_F_ATTR(i) BIT(i) 35 36 struct seg6_local_lwt; 37 38 /* callbacks used for customizing the creation and destruction of a behavior */ 39 struct seg6_local_lwtunnel_ops { 40 int (*build_state)(struct seg6_local_lwt *slwt, const void *cfg, 41 struct netlink_ext_ack *extack); 42 void (*destroy_state)(struct seg6_local_lwt *slwt); 43 }; 44 45 struct seg6_action_desc { 46 int action; 47 unsigned long attrs; 48 49 /* The optattrs field is used for specifying all the optional 50 * attributes supported by a specific behavior. 51 * It means that if one of these attributes is not provided in the 52 * netlink message during the behavior creation, no errors will be 53 * returned to the userspace. 54 * 55 * Each attribute can be only of two types (mutually exclusive): 56 * 1) required or 2) optional. 57 * Every user MUST obey to this rule! If you set an attribute as 58 * required the same attribute CANNOT be set as optional and vice 59 * versa. 60 */ 61 unsigned long optattrs; 62 63 int (*input)(struct sk_buff *skb, struct seg6_local_lwt *slwt); 64 int static_headroom; 65 66 struct seg6_local_lwtunnel_ops slwt_ops; 67 }; 68 69 struct bpf_lwt_prog { 70 struct bpf_prog *prog; 71 char *name; 72 }; 73 74 enum seg6_end_dt_mode { 75 DT_INVALID_MODE = -EINVAL, 76 DT_LEGACY_MODE = 0, 77 DT_VRF_MODE = 1, 78 }; 79 80 struct seg6_end_dt_info { 81 enum seg6_end_dt_mode mode; 82 83 struct net *net; 84 /* VRF device associated to the routing table used by the SRv6 85 * End.DT4/DT6 behavior for routing IPv4/IPv6 packets. 86 */ 87 int vrf_ifindex; 88 int vrf_table; 89 90 /* tunneled packet family (IPv4 or IPv6). 91 * Protocol and header length are inferred from family. 92 */ 93 u16 family; 94 }; 95 96 struct pcpu_seg6_local_counters { 97 u64_stats_t packets; 98 u64_stats_t bytes; 99 u64_stats_t errors; 100 101 struct u64_stats_sync syncp; 102 }; 103 104 /* This struct groups all the SRv6 Behavior counters supported so far. 105 * 106 * put_nla_counters() makes use of this data structure to collect all counter 107 * values after the per-CPU counter evaluation has been performed. 108 * Finally, each counter value (in seg6_local_counters) is stored in the 109 * corresponding netlink attribute and sent to user space. 110 * 111 * NB: we don't want to expose this structure to user space! 112 */ 113 struct seg6_local_counters { 114 __u64 packets; 115 __u64 bytes; 116 __u64 errors; 117 }; 118 119 #define seg6_local_alloc_pcpu_counters(__gfp) \ 120 __netdev_alloc_pcpu_stats(struct pcpu_seg6_local_counters, \ 121 ((__gfp) | __GFP_ZERO)) 122 123 #define SEG6_F_LOCAL_COUNTERS SEG6_F_ATTR(SEG6_LOCAL_COUNTERS) 124 125 struct seg6_local_lwt { 126 int action; 127 struct ipv6_sr_hdr *srh; 128 int table; 129 struct in_addr nh4; 130 struct in6_addr nh6; 131 int iif; 132 int oif; 133 struct bpf_lwt_prog bpf; 134 #ifdef CONFIG_NET_L3_MASTER_DEV 135 struct seg6_end_dt_info dt_info; 136 #endif 137 struct pcpu_seg6_local_counters __percpu *pcpu_counters; 138 139 int headroom; 140 struct seg6_action_desc *desc; 141 /* unlike the required attrs, we have to track the optional attributes 142 * that have been effectively parsed. 143 */ 144 unsigned long parsed_optattrs; 145 }; 146 147 static struct seg6_local_lwt *seg6_local_lwtunnel(struct lwtunnel_state *lwt) 148 { 149 return (struct seg6_local_lwt *)lwt->data; 150 } 151 152 static struct ipv6_sr_hdr *get_srh(struct sk_buff *skb, int flags) 153 { 154 struct ipv6_sr_hdr *srh; 155 int len, srhoff = 0; 156 157 if (ipv6_find_hdr(skb, &srhoff, IPPROTO_ROUTING, NULL, &flags) < 0) 158 return NULL; 159 160 if (!pskb_may_pull(skb, srhoff + sizeof(*srh))) 161 return NULL; 162 163 srh = (struct ipv6_sr_hdr *)(skb->data + srhoff); 164 165 len = (srh->hdrlen + 1) << 3; 166 167 if (!pskb_may_pull(skb, srhoff + len)) 168 return NULL; 169 170 /* note that pskb_may_pull may change pointers in header; 171 * for this reason it is necessary to reload them when needed. 172 */ 173 srh = (struct ipv6_sr_hdr *)(skb->data + srhoff); 174 175 if (!seg6_validate_srh(srh, len, true)) 176 return NULL; 177 178 return srh; 179 } 180 181 static struct ipv6_sr_hdr *get_and_validate_srh(struct sk_buff *skb) 182 { 183 struct ipv6_sr_hdr *srh; 184 185 srh = get_srh(skb, IP6_FH_F_SKIP_RH); 186 if (!srh) 187 return NULL; 188 189 #ifdef CONFIG_IPV6_SEG6_HMAC 190 if (!seg6_hmac_validate_skb(skb)) 191 return NULL; 192 #endif 193 194 return srh; 195 } 196 197 static bool decap_and_validate(struct sk_buff *skb, int proto) 198 { 199 struct ipv6_sr_hdr *srh; 200 unsigned int off = 0; 201 202 srh = get_srh(skb, 0); 203 if (srh && srh->segments_left > 0) 204 return false; 205 206 #ifdef CONFIG_IPV6_SEG6_HMAC 207 if (srh && !seg6_hmac_validate_skb(skb)) 208 return false; 209 #endif 210 211 if (ipv6_find_hdr(skb, &off, proto, NULL, NULL) < 0) 212 return false; 213 214 if (!pskb_pull(skb, off)) 215 return false; 216 217 skb_postpull_rcsum(skb, skb_network_header(skb), off); 218 219 skb_reset_network_header(skb); 220 skb_reset_transport_header(skb); 221 if (iptunnel_pull_offloads(skb)) 222 return false; 223 224 return true; 225 } 226 227 static void advance_nextseg(struct ipv6_sr_hdr *srh, struct in6_addr *daddr) 228 { 229 struct in6_addr *addr; 230 231 srh->segments_left--; 232 addr = srh->segments + srh->segments_left; 233 *daddr = *addr; 234 } 235 236 static int 237 seg6_lookup_any_nexthop(struct sk_buff *skb, struct in6_addr *nhaddr, 238 u32 tbl_id, bool local_delivery) 239 { 240 struct net *net = dev_net(skb->dev); 241 struct ipv6hdr *hdr = ipv6_hdr(skb); 242 int flags = RT6_LOOKUP_F_HAS_SADDR; 243 struct dst_entry *dst = NULL; 244 struct rt6_info *rt; 245 struct flowi6 fl6; 246 int dev_flags = 0; 247 248 fl6.flowi6_iif = skb->dev->ifindex; 249 fl6.daddr = nhaddr ? *nhaddr : hdr->daddr; 250 fl6.saddr = hdr->saddr; 251 fl6.flowlabel = ip6_flowinfo(hdr); 252 fl6.flowi6_mark = skb->mark; 253 fl6.flowi6_proto = hdr->nexthdr; 254 255 if (nhaddr) 256 fl6.flowi6_flags = FLOWI_FLAG_KNOWN_NH; 257 258 if (!tbl_id) { 259 dst = ip6_route_input_lookup(net, skb->dev, &fl6, skb, flags); 260 } else { 261 struct fib6_table *table; 262 263 table = fib6_get_table(net, tbl_id); 264 if (!table) 265 goto out; 266 267 rt = ip6_pol_route(net, table, 0, &fl6, skb, flags); 268 dst = &rt->dst; 269 } 270 271 /* we want to discard traffic destined for local packet processing, 272 * if @local_delivery is set to false. 273 */ 274 if (!local_delivery) 275 dev_flags |= IFF_LOOPBACK; 276 277 if (dst && (dst->dev->flags & dev_flags) && !dst->error) { 278 dst_release(dst); 279 dst = NULL; 280 } 281 282 out: 283 if (!dst) { 284 rt = net->ipv6.ip6_blk_hole_entry; 285 dst = &rt->dst; 286 dst_hold(dst); 287 } 288 289 skb_dst_drop(skb); 290 skb_dst_set(skb, dst); 291 return dst->error; 292 } 293 294 int seg6_lookup_nexthop(struct sk_buff *skb, 295 struct in6_addr *nhaddr, u32 tbl_id) 296 { 297 return seg6_lookup_any_nexthop(skb, nhaddr, tbl_id, false); 298 } 299 300 /* regular endpoint function */ 301 static int input_action_end(struct sk_buff *skb, struct seg6_local_lwt *slwt) 302 { 303 struct ipv6_sr_hdr *srh; 304 305 srh = get_and_validate_srh(skb); 306 if (!srh) 307 goto drop; 308 309 advance_nextseg(srh, &ipv6_hdr(skb)->daddr); 310 311 seg6_lookup_nexthop(skb, NULL, 0); 312 313 return dst_input(skb); 314 315 drop: 316 kfree_skb(skb); 317 return -EINVAL; 318 } 319 320 /* regular endpoint, and forward to specified nexthop */ 321 static int input_action_end_x(struct sk_buff *skb, struct seg6_local_lwt *slwt) 322 { 323 struct ipv6_sr_hdr *srh; 324 325 srh = get_and_validate_srh(skb); 326 if (!srh) 327 goto drop; 328 329 advance_nextseg(srh, &ipv6_hdr(skb)->daddr); 330 331 seg6_lookup_nexthop(skb, &slwt->nh6, 0); 332 333 return dst_input(skb); 334 335 drop: 336 kfree_skb(skb); 337 return -EINVAL; 338 } 339 340 static int input_action_end_t(struct sk_buff *skb, struct seg6_local_lwt *slwt) 341 { 342 struct ipv6_sr_hdr *srh; 343 344 srh = get_and_validate_srh(skb); 345 if (!srh) 346 goto drop; 347 348 advance_nextseg(srh, &ipv6_hdr(skb)->daddr); 349 350 seg6_lookup_nexthop(skb, NULL, slwt->table); 351 352 return dst_input(skb); 353 354 drop: 355 kfree_skb(skb); 356 return -EINVAL; 357 } 358 359 /* decapsulate and forward inner L2 frame on specified interface */ 360 static int input_action_end_dx2(struct sk_buff *skb, 361 struct seg6_local_lwt *slwt) 362 { 363 struct net *net = dev_net(skb->dev); 364 struct net_device *odev; 365 struct ethhdr *eth; 366 367 if (!decap_and_validate(skb, IPPROTO_ETHERNET)) 368 goto drop; 369 370 if (!pskb_may_pull(skb, ETH_HLEN)) 371 goto drop; 372 373 skb_reset_mac_header(skb); 374 eth = (struct ethhdr *)skb->data; 375 376 /* To determine the frame's protocol, we assume it is 802.3. This avoids 377 * a call to eth_type_trans(), which is not really relevant for our 378 * use case. 379 */ 380 if (!eth_proto_is_802_3(eth->h_proto)) 381 goto drop; 382 383 odev = dev_get_by_index_rcu(net, slwt->oif); 384 if (!odev) 385 goto drop; 386 387 /* As we accept Ethernet frames, make sure the egress device is of 388 * the correct type. 389 */ 390 if (odev->type != ARPHRD_ETHER) 391 goto drop; 392 393 if (!(odev->flags & IFF_UP) || !netif_carrier_ok(odev)) 394 goto drop; 395 396 skb_orphan(skb); 397 398 if (skb_warn_if_lro(skb)) 399 goto drop; 400 401 skb_forward_csum(skb); 402 403 if (skb->len - ETH_HLEN > odev->mtu) 404 goto drop; 405 406 skb->dev = odev; 407 skb->protocol = eth->h_proto; 408 409 return dev_queue_xmit(skb); 410 411 drop: 412 kfree_skb(skb); 413 return -EINVAL; 414 } 415 416 /* decapsulate and forward to specified nexthop */ 417 static int input_action_end_dx6(struct sk_buff *skb, 418 struct seg6_local_lwt *slwt) 419 { 420 struct in6_addr *nhaddr = NULL; 421 422 /* this function accepts IPv6 encapsulated packets, with either 423 * an SRH with SL=0, or no SRH. 424 */ 425 426 if (!decap_and_validate(skb, IPPROTO_IPV6)) 427 goto drop; 428 429 if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) 430 goto drop; 431 432 /* The inner packet is not associated to any local interface, 433 * so we do not call netif_rx(). 434 * 435 * If slwt->nh6 is set to ::, then lookup the nexthop for the 436 * inner packet's DA. Otherwise, use the specified nexthop. 437 */ 438 439 if (!ipv6_addr_any(&slwt->nh6)) 440 nhaddr = &slwt->nh6; 441 442 skb_set_transport_header(skb, sizeof(struct ipv6hdr)); 443 444 seg6_lookup_nexthop(skb, nhaddr, 0); 445 446 return dst_input(skb); 447 drop: 448 kfree_skb(skb); 449 return -EINVAL; 450 } 451 452 static int input_action_end_dx4(struct sk_buff *skb, 453 struct seg6_local_lwt *slwt) 454 { 455 struct iphdr *iph; 456 __be32 nhaddr; 457 int err; 458 459 if (!decap_and_validate(skb, IPPROTO_IPIP)) 460 goto drop; 461 462 if (!pskb_may_pull(skb, sizeof(struct iphdr))) 463 goto drop; 464 465 skb->protocol = htons(ETH_P_IP); 466 467 iph = ip_hdr(skb); 468 469 nhaddr = slwt->nh4.s_addr ?: iph->daddr; 470 471 skb_dst_drop(skb); 472 473 skb_set_transport_header(skb, sizeof(struct iphdr)); 474 475 err = ip_route_input(skb, nhaddr, iph->saddr, 0, skb->dev); 476 if (err) 477 goto drop; 478 479 return dst_input(skb); 480 481 drop: 482 kfree_skb(skb); 483 return -EINVAL; 484 } 485 486 #ifdef CONFIG_NET_L3_MASTER_DEV 487 static struct net *fib6_config_get_net(const struct fib6_config *fib6_cfg) 488 { 489 const struct nl_info *nli = &fib6_cfg->fc_nlinfo; 490 491 return nli->nl_net; 492 } 493 494 static int __seg6_end_dt_vrf_build(struct seg6_local_lwt *slwt, const void *cfg, 495 u16 family, struct netlink_ext_ack *extack) 496 { 497 struct seg6_end_dt_info *info = &slwt->dt_info; 498 int vrf_ifindex; 499 struct net *net; 500 501 net = fib6_config_get_net(cfg); 502 503 /* note that vrf_table was already set by parse_nla_vrftable() */ 504 vrf_ifindex = l3mdev_ifindex_lookup_by_table_id(L3MDEV_TYPE_VRF, net, 505 info->vrf_table); 506 if (vrf_ifindex < 0) { 507 if (vrf_ifindex == -EPERM) { 508 NL_SET_ERR_MSG(extack, 509 "Strict mode for VRF is disabled"); 510 } else if (vrf_ifindex == -ENODEV) { 511 NL_SET_ERR_MSG(extack, 512 "Table has no associated VRF device"); 513 } else { 514 pr_debug("seg6local: SRv6 End.DT* creation error=%d\n", 515 vrf_ifindex); 516 } 517 518 return vrf_ifindex; 519 } 520 521 info->net = net; 522 info->vrf_ifindex = vrf_ifindex; 523 524 info->family = family; 525 info->mode = DT_VRF_MODE; 526 527 return 0; 528 } 529 530 /* The SRv6 End.DT4/DT6 behavior extracts the inner (IPv4/IPv6) packet and 531 * routes the IPv4/IPv6 packet by looking at the configured routing table. 532 * 533 * In the SRv6 End.DT4/DT6 use case, we can receive traffic (IPv6+Segment 534 * Routing Header packets) from several interfaces and the outer IPv6 535 * destination address (DA) is used for retrieving the specific instance of the 536 * End.DT4/DT6 behavior that should process the packets. 537 * 538 * However, the inner IPv4/IPv6 packet is not really bound to any receiving 539 * interface and thus the End.DT4/DT6 sets the VRF (associated with the 540 * corresponding routing table) as the *receiving* interface. 541 * In other words, the End.DT4/DT6 processes a packet as if it has been received 542 * directly by the VRF (and not by one of its slave devices, if any). 543 * In this way, the VRF interface is used for routing the IPv4/IPv6 packet in 544 * according to the routing table configured by the End.DT4/DT6 instance. 545 * 546 * This design allows you to get some interesting features like: 547 * 1) the statistics on rx packets; 548 * 2) the possibility to install a packet sniffer on the receiving interface 549 * (the VRF one) for looking at the incoming packets; 550 * 3) the possibility to leverage the netfilter prerouting hook for the inner 551 * IPv4 packet. 552 * 553 * This function returns: 554 * - the sk_buff* when the VRF rcv handler has processed the packet correctly; 555 * - NULL when the skb is consumed by the VRF rcv handler; 556 * - a pointer which encodes a negative error number in case of error. 557 * Note that in this case, the function takes care of freeing the skb. 558 */ 559 static struct sk_buff *end_dt_vrf_rcv(struct sk_buff *skb, u16 family, 560 struct net_device *dev) 561 { 562 /* based on l3mdev_ip_rcv; we are only interested in the master */ 563 if (unlikely(!netif_is_l3_master(dev) && !netif_has_l3_rx_handler(dev))) 564 goto drop; 565 566 if (unlikely(!dev->l3mdev_ops->l3mdev_l3_rcv)) 567 goto drop; 568 569 /* the decap packet IPv4/IPv6 does not come with any mac header info. 570 * We must unset the mac header to allow the VRF device to rebuild it, 571 * just in case there is a sniffer attached on the device. 572 */ 573 skb_unset_mac_header(skb); 574 575 skb = dev->l3mdev_ops->l3mdev_l3_rcv(dev, skb, family); 576 if (!skb) 577 /* the skb buffer was consumed by the handler */ 578 return NULL; 579 580 /* when a packet is received by a VRF or by one of its slaves, the 581 * master device reference is set into the skb. 582 */ 583 if (unlikely(skb->dev != dev || skb->skb_iif != dev->ifindex)) 584 goto drop; 585 586 return skb; 587 588 drop: 589 kfree_skb(skb); 590 return ERR_PTR(-EINVAL); 591 } 592 593 static struct net_device *end_dt_get_vrf_rcu(struct sk_buff *skb, 594 struct seg6_end_dt_info *info) 595 { 596 int vrf_ifindex = info->vrf_ifindex; 597 struct net *net = info->net; 598 599 if (unlikely(vrf_ifindex < 0)) 600 goto error; 601 602 if (unlikely(!net_eq(dev_net(skb->dev), net))) 603 goto error; 604 605 return dev_get_by_index_rcu(net, vrf_ifindex); 606 607 error: 608 return NULL; 609 } 610 611 static struct sk_buff *end_dt_vrf_core(struct sk_buff *skb, 612 struct seg6_local_lwt *slwt, u16 family) 613 { 614 struct seg6_end_dt_info *info = &slwt->dt_info; 615 struct net_device *vrf; 616 __be16 protocol; 617 int hdrlen; 618 619 vrf = end_dt_get_vrf_rcu(skb, info); 620 if (unlikely(!vrf)) 621 goto drop; 622 623 switch (family) { 624 case AF_INET: 625 protocol = htons(ETH_P_IP); 626 hdrlen = sizeof(struct iphdr); 627 break; 628 case AF_INET6: 629 protocol = htons(ETH_P_IPV6); 630 hdrlen = sizeof(struct ipv6hdr); 631 break; 632 case AF_UNSPEC: 633 fallthrough; 634 default: 635 goto drop; 636 } 637 638 if (unlikely(info->family != AF_UNSPEC && info->family != family)) { 639 pr_warn_once("seg6local: SRv6 End.DT* family mismatch"); 640 goto drop; 641 } 642 643 skb->protocol = protocol; 644 645 skb_dst_drop(skb); 646 647 skb_set_transport_header(skb, hdrlen); 648 649 return end_dt_vrf_rcv(skb, family, vrf); 650 651 drop: 652 kfree_skb(skb); 653 return ERR_PTR(-EINVAL); 654 } 655 656 static int input_action_end_dt4(struct sk_buff *skb, 657 struct seg6_local_lwt *slwt) 658 { 659 struct iphdr *iph; 660 int err; 661 662 if (!decap_and_validate(skb, IPPROTO_IPIP)) 663 goto drop; 664 665 if (!pskb_may_pull(skb, sizeof(struct iphdr))) 666 goto drop; 667 668 skb = end_dt_vrf_core(skb, slwt, AF_INET); 669 if (!skb) 670 /* packet has been processed and consumed by the VRF */ 671 return 0; 672 673 if (IS_ERR(skb)) 674 return PTR_ERR(skb); 675 676 iph = ip_hdr(skb); 677 678 err = ip_route_input(skb, iph->daddr, iph->saddr, 0, skb->dev); 679 if (unlikely(err)) 680 goto drop; 681 682 return dst_input(skb); 683 684 drop: 685 kfree_skb(skb); 686 return -EINVAL; 687 } 688 689 static int seg6_end_dt4_build(struct seg6_local_lwt *slwt, const void *cfg, 690 struct netlink_ext_ack *extack) 691 { 692 return __seg6_end_dt_vrf_build(slwt, cfg, AF_INET, extack); 693 } 694 695 static enum 696 seg6_end_dt_mode seg6_end_dt6_parse_mode(struct seg6_local_lwt *slwt) 697 { 698 unsigned long parsed_optattrs = slwt->parsed_optattrs; 699 bool legacy, vrfmode; 700 701 legacy = !!(parsed_optattrs & SEG6_F_ATTR(SEG6_LOCAL_TABLE)); 702 vrfmode = !!(parsed_optattrs & SEG6_F_ATTR(SEG6_LOCAL_VRFTABLE)); 703 704 if (!(legacy ^ vrfmode)) 705 /* both are absent or present: invalid DT6 mode */ 706 return DT_INVALID_MODE; 707 708 return legacy ? DT_LEGACY_MODE : DT_VRF_MODE; 709 } 710 711 static enum seg6_end_dt_mode seg6_end_dt6_get_mode(struct seg6_local_lwt *slwt) 712 { 713 struct seg6_end_dt_info *info = &slwt->dt_info; 714 715 return info->mode; 716 } 717 718 static int seg6_end_dt6_build(struct seg6_local_lwt *slwt, const void *cfg, 719 struct netlink_ext_ack *extack) 720 { 721 enum seg6_end_dt_mode mode = seg6_end_dt6_parse_mode(slwt); 722 struct seg6_end_dt_info *info = &slwt->dt_info; 723 724 switch (mode) { 725 case DT_LEGACY_MODE: 726 info->mode = DT_LEGACY_MODE; 727 return 0; 728 case DT_VRF_MODE: 729 return __seg6_end_dt_vrf_build(slwt, cfg, AF_INET6, extack); 730 default: 731 NL_SET_ERR_MSG(extack, "table or vrftable must be specified"); 732 return -EINVAL; 733 } 734 } 735 #endif 736 737 static int input_action_end_dt6(struct sk_buff *skb, 738 struct seg6_local_lwt *slwt) 739 { 740 if (!decap_and_validate(skb, IPPROTO_IPV6)) 741 goto drop; 742 743 if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) 744 goto drop; 745 746 #ifdef CONFIG_NET_L3_MASTER_DEV 747 if (seg6_end_dt6_get_mode(slwt) == DT_LEGACY_MODE) 748 goto legacy_mode; 749 750 /* DT6_VRF_MODE */ 751 skb = end_dt_vrf_core(skb, slwt, AF_INET6); 752 if (!skb) 753 /* packet has been processed and consumed by the VRF */ 754 return 0; 755 756 if (IS_ERR(skb)) 757 return PTR_ERR(skb); 758 759 /* note: this time we do not need to specify the table because the VRF 760 * takes care of selecting the correct table. 761 */ 762 seg6_lookup_any_nexthop(skb, NULL, 0, true); 763 764 return dst_input(skb); 765 766 legacy_mode: 767 #endif 768 skb_set_transport_header(skb, sizeof(struct ipv6hdr)); 769 770 seg6_lookup_any_nexthop(skb, NULL, slwt->table, true); 771 772 return dst_input(skb); 773 774 drop: 775 kfree_skb(skb); 776 return -EINVAL; 777 } 778 779 #ifdef CONFIG_NET_L3_MASTER_DEV 780 static int seg6_end_dt46_build(struct seg6_local_lwt *slwt, const void *cfg, 781 struct netlink_ext_ack *extack) 782 { 783 return __seg6_end_dt_vrf_build(slwt, cfg, AF_UNSPEC, extack); 784 } 785 786 static int input_action_end_dt46(struct sk_buff *skb, 787 struct seg6_local_lwt *slwt) 788 { 789 unsigned int off = 0; 790 int nexthdr; 791 792 nexthdr = ipv6_find_hdr(skb, &off, -1, NULL, NULL); 793 if (unlikely(nexthdr < 0)) 794 goto drop; 795 796 switch (nexthdr) { 797 case IPPROTO_IPIP: 798 return input_action_end_dt4(skb, slwt); 799 case IPPROTO_IPV6: 800 return input_action_end_dt6(skb, slwt); 801 } 802 803 drop: 804 kfree_skb(skb); 805 return -EINVAL; 806 } 807 #endif 808 809 /* push an SRH on top of the current one */ 810 static int input_action_end_b6(struct sk_buff *skb, struct seg6_local_lwt *slwt) 811 { 812 struct ipv6_sr_hdr *srh; 813 int err = -EINVAL; 814 815 srh = get_and_validate_srh(skb); 816 if (!srh) 817 goto drop; 818 819 err = seg6_do_srh_inline(skb, slwt->srh); 820 if (err) 821 goto drop; 822 823 ipv6_hdr(skb)->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); 824 skb_set_transport_header(skb, sizeof(struct ipv6hdr)); 825 826 seg6_lookup_nexthop(skb, NULL, 0); 827 828 return dst_input(skb); 829 830 drop: 831 kfree_skb(skb); 832 return err; 833 } 834 835 /* encapsulate within an outer IPv6 header and a specified SRH */ 836 static int input_action_end_b6_encap(struct sk_buff *skb, 837 struct seg6_local_lwt *slwt) 838 { 839 struct ipv6_sr_hdr *srh; 840 int err = -EINVAL; 841 842 srh = get_and_validate_srh(skb); 843 if (!srh) 844 goto drop; 845 846 advance_nextseg(srh, &ipv6_hdr(skb)->daddr); 847 848 skb_reset_inner_headers(skb); 849 skb->encapsulation = 1; 850 851 err = seg6_do_srh_encap(skb, slwt->srh, IPPROTO_IPV6); 852 if (err) 853 goto drop; 854 855 ipv6_hdr(skb)->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); 856 skb_set_transport_header(skb, sizeof(struct ipv6hdr)); 857 858 seg6_lookup_nexthop(skb, NULL, 0); 859 860 return dst_input(skb); 861 862 drop: 863 kfree_skb(skb); 864 return err; 865 } 866 867 DEFINE_PER_CPU(struct seg6_bpf_srh_state, seg6_bpf_srh_states); 868 869 bool seg6_bpf_has_valid_srh(struct sk_buff *skb) 870 { 871 struct seg6_bpf_srh_state *srh_state = 872 this_cpu_ptr(&seg6_bpf_srh_states); 873 struct ipv6_sr_hdr *srh = srh_state->srh; 874 875 if (unlikely(srh == NULL)) 876 return false; 877 878 if (unlikely(!srh_state->valid)) { 879 if ((srh_state->hdrlen & 7) != 0) 880 return false; 881 882 srh->hdrlen = (u8)(srh_state->hdrlen >> 3); 883 if (!seg6_validate_srh(srh, (srh->hdrlen + 1) << 3, true)) 884 return false; 885 886 srh_state->valid = true; 887 } 888 889 return true; 890 } 891 892 static int input_action_end_bpf(struct sk_buff *skb, 893 struct seg6_local_lwt *slwt) 894 { 895 struct seg6_bpf_srh_state *srh_state = 896 this_cpu_ptr(&seg6_bpf_srh_states); 897 struct ipv6_sr_hdr *srh; 898 int ret; 899 900 srh = get_and_validate_srh(skb); 901 if (!srh) { 902 kfree_skb(skb); 903 return -EINVAL; 904 } 905 advance_nextseg(srh, &ipv6_hdr(skb)->daddr); 906 907 /* preempt_disable is needed to protect the per-CPU buffer srh_state, 908 * which is also accessed by the bpf_lwt_seg6_* helpers 909 */ 910 preempt_disable(); 911 srh_state->srh = srh; 912 srh_state->hdrlen = srh->hdrlen << 3; 913 srh_state->valid = true; 914 915 rcu_read_lock(); 916 bpf_compute_data_pointers(skb); 917 ret = bpf_prog_run_save_cb(slwt->bpf.prog, skb); 918 rcu_read_unlock(); 919 920 switch (ret) { 921 case BPF_OK: 922 case BPF_REDIRECT: 923 break; 924 case BPF_DROP: 925 goto drop; 926 default: 927 pr_warn_once("bpf-seg6local: Illegal return value %u\n", ret); 928 goto drop; 929 } 930 931 if (srh_state->srh && !seg6_bpf_has_valid_srh(skb)) 932 goto drop; 933 934 preempt_enable(); 935 if (ret != BPF_REDIRECT) 936 seg6_lookup_nexthop(skb, NULL, 0); 937 938 return dst_input(skb); 939 940 drop: 941 preempt_enable(); 942 kfree_skb(skb); 943 return -EINVAL; 944 } 945 946 static struct seg6_action_desc seg6_action_table[] = { 947 { 948 .action = SEG6_LOCAL_ACTION_END, 949 .attrs = 0, 950 .optattrs = SEG6_F_LOCAL_COUNTERS, 951 .input = input_action_end, 952 }, 953 { 954 .action = SEG6_LOCAL_ACTION_END_X, 955 .attrs = SEG6_F_ATTR(SEG6_LOCAL_NH6), 956 .optattrs = SEG6_F_LOCAL_COUNTERS, 957 .input = input_action_end_x, 958 }, 959 { 960 .action = SEG6_LOCAL_ACTION_END_T, 961 .attrs = SEG6_F_ATTR(SEG6_LOCAL_TABLE), 962 .optattrs = SEG6_F_LOCAL_COUNTERS, 963 .input = input_action_end_t, 964 }, 965 { 966 .action = SEG6_LOCAL_ACTION_END_DX2, 967 .attrs = SEG6_F_ATTR(SEG6_LOCAL_OIF), 968 .optattrs = SEG6_F_LOCAL_COUNTERS, 969 .input = input_action_end_dx2, 970 }, 971 { 972 .action = SEG6_LOCAL_ACTION_END_DX6, 973 .attrs = SEG6_F_ATTR(SEG6_LOCAL_NH6), 974 .optattrs = SEG6_F_LOCAL_COUNTERS, 975 .input = input_action_end_dx6, 976 }, 977 { 978 .action = SEG6_LOCAL_ACTION_END_DX4, 979 .attrs = SEG6_F_ATTR(SEG6_LOCAL_NH4), 980 .optattrs = SEG6_F_LOCAL_COUNTERS, 981 .input = input_action_end_dx4, 982 }, 983 { 984 .action = SEG6_LOCAL_ACTION_END_DT4, 985 .attrs = SEG6_F_ATTR(SEG6_LOCAL_VRFTABLE), 986 .optattrs = SEG6_F_LOCAL_COUNTERS, 987 #ifdef CONFIG_NET_L3_MASTER_DEV 988 .input = input_action_end_dt4, 989 .slwt_ops = { 990 .build_state = seg6_end_dt4_build, 991 }, 992 #endif 993 }, 994 { 995 .action = SEG6_LOCAL_ACTION_END_DT6, 996 #ifdef CONFIG_NET_L3_MASTER_DEV 997 .attrs = 0, 998 .optattrs = SEG6_F_LOCAL_COUNTERS | 999 SEG6_F_ATTR(SEG6_LOCAL_TABLE) | 1000 SEG6_F_ATTR(SEG6_LOCAL_VRFTABLE), 1001 .slwt_ops = { 1002 .build_state = seg6_end_dt6_build, 1003 }, 1004 #else 1005 .attrs = SEG6_F_ATTR(SEG6_LOCAL_TABLE), 1006 .optattrs = SEG6_F_LOCAL_COUNTERS, 1007 #endif 1008 .input = input_action_end_dt6, 1009 }, 1010 { 1011 .action = SEG6_LOCAL_ACTION_END_DT46, 1012 .attrs = SEG6_F_ATTR(SEG6_LOCAL_VRFTABLE), 1013 .optattrs = SEG6_F_LOCAL_COUNTERS, 1014 #ifdef CONFIG_NET_L3_MASTER_DEV 1015 .input = input_action_end_dt46, 1016 .slwt_ops = { 1017 .build_state = seg6_end_dt46_build, 1018 }, 1019 #endif 1020 }, 1021 { 1022 .action = SEG6_LOCAL_ACTION_END_B6, 1023 .attrs = SEG6_F_ATTR(SEG6_LOCAL_SRH), 1024 .optattrs = SEG6_F_LOCAL_COUNTERS, 1025 .input = input_action_end_b6, 1026 }, 1027 { 1028 .action = SEG6_LOCAL_ACTION_END_B6_ENCAP, 1029 .attrs = SEG6_F_ATTR(SEG6_LOCAL_SRH), 1030 .optattrs = SEG6_F_LOCAL_COUNTERS, 1031 .input = input_action_end_b6_encap, 1032 .static_headroom = sizeof(struct ipv6hdr), 1033 }, 1034 { 1035 .action = SEG6_LOCAL_ACTION_END_BPF, 1036 .attrs = SEG6_F_ATTR(SEG6_LOCAL_BPF), 1037 .optattrs = SEG6_F_LOCAL_COUNTERS, 1038 .input = input_action_end_bpf, 1039 }, 1040 1041 }; 1042 1043 static struct seg6_action_desc *__get_action_desc(int action) 1044 { 1045 struct seg6_action_desc *desc; 1046 int i, count; 1047 1048 count = ARRAY_SIZE(seg6_action_table); 1049 for (i = 0; i < count; i++) { 1050 desc = &seg6_action_table[i]; 1051 if (desc->action == action) 1052 return desc; 1053 } 1054 1055 return NULL; 1056 } 1057 1058 static bool seg6_lwtunnel_counters_enabled(struct seg6_local_lwt *slwt) 1059 { 1060 return slwt->parsed_optattrs & SEG6_F_LOCAL_COUNTERS; 1061 } 1062 1063 static void seg6_local_update_counters(struct seg6_local_lwt *slwt, 1064 unsigned int len, int err) 1065 { 1066 struct pcpu_seg6_local_counters *pcounters; 1067 1068 pcounters = this_cpu_ptr(slwt->pcpu_counters); 1069 u64_stats_update_begin(&pcounters->syncp); 1070 1071 if (likely(!err)) { 1072 u64_stats_inc(&pcounters->packets); 1073 u64_stats_add(&pcounters->bytes, len); 1074 } else { 1075 u64_stats_inc(&pcounters->errors); 1076 } 1077 1078 u64_stats_update_end(&pcounters->syncp); 1079 } 1080 1081 static int seg6_local_input(struct sk_buff *skb) 1082 { 1083 struct dst_entry *orig_dst = skb_dst(skb); 1084 struct seg6_action_desc *desc; 1085 struct seg6_local_lwt *slwt; 1086 unsigned int len = skb->len; 1087 int rc; 1088 1089 if (skb->protocol != htons(ETH_P_IPV6)) { 1090 kfree_skb(skb); 1091 return -EINVAL; 1092 } 1093 1094 slwt = seg6_local_lwtunnel(orig_dst->lwtstate); 1095 desc = slwt->desc; 1096 1097 rc = desc->input(skb, slwt); 1098 1099 if (!seg6_lwtunnel_counters_enabled(slwt)) 1100 return rc; 1101 1102 seg6_local_update_counters(slwt, len, rc); 1103 1104 return rc; 1105 } 1106 1107 static const struct nla_policy seg6_local_policy[SEG6_LOCAL_MAX + 1] = { 1108 [SEG6_LOCAL_ACTION] = { .type = NLA_U32 }, 1109 [SEG6_LOCAL_SRH] = { .type = NLA_BINARY }, 1110 [SEG6_LOCAL_TABLE] = { .type = NLA_U32 }, 1111 [SEG6_LOCAL_VRFTABLE] = { .type = NLA_U32 }, 1112 [SEG6_LOCAL_NH4] = { .type = NLA_BINARY, 1113 .len = sizeof(struct in_addr) }, 1114 [SEG6_LOCAL_NH6] = { .type = NLA_BINARY, 1115 .len = sizeof(struct in6_addr) }, 1116 [SEG6_LOCAL_IIF] = { .type = NLA_U32 }, 1117 [SEG6_LOCAL_OIF] = { .type = NLA_U32 }, 1118 [SEG6_LOCAL_BPF] = { .type = NLA_NESTED }, 1119 [SEG6_LOCAL_COUNTERS] = { .type = NLA_NESTED }, 1120 }; 1121 1122 static int parse_nla_srh(struct nlattr **attrs, struct seg6_local_lwt *slwt) 1123 { 1124 struct ipv6_sr_hdr *srh; 1125 int len; 1126 1127 srh = nla_data(attrs[SEG6_LOCAL_SRH]); 1128 len = nla_len(attrs[SEG6_LOCAL_SRH]); 1129 1130 /* SRH must contain at least one segment */ 1131 if (len < sizeof(*srh) + sizeof(struct in6_addr)) 1132 return -EINVAL; 1133 1134 if (!seg6_validate_srh(srh, len, false)) 1135 return -EINVAL; 1136 1137 slwt->srh = kmemdup(srh, len, GFP_KERNEL); 1138 if (!slwt->srh) 1139 return -ENOMEM; 1140 1141 slwt->headroom += len; 1142 1143 return 0; 1144 } 1145 1146 static int put_nla_srh(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1147 { 1148 struct ipv6_sr_hdr *srh; 1149 struct nlattr *nla; 1150 int len; 1151 1152 srh = slwt->srh; 1153 len = (srh->hdrlen + 1) << 3; 1154 1155 nla = nla_reserve(skb, SEG6_LOCAL_SRH, len); 1156 if (!nla) 1157 return -EMSGSIZE; 1158 1159 memcpy(nla_data(nla), srh, len); 1160 1161 return 0; 1162 } 1163 1164 static int cmp_nla_srh(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1165 { 1166 int len = (a->srh->hdrlen + 1) << 3; 1167 1168 if (len != ((b->srh->hdrlen + 1) << 3)) 1169 return 1; 1170 1171 return memcmp(a->srh, b->srh, len); 1172 } 1173 1174 static void destroy_attr_srh(struct seg6_local_lwt *slwt) 1175 { 1176 kfree(slwt->srh); 1177 } 1178 1179 static int parse_nla_table(struct nlattr **attrs, struct seg6_local_lwt *slwt) 1180 { 1181 slwt->table = nla_get_u32(attrs[SEG6_LOCAL_TABLE]); 1182 1183 return 0; 1184 } 1185 1186 static int put_nla_table(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1187 { 1188 if (nla_put_u32(skb, SEG6_LOCAL_TABLE, slwt->table)) 1189 return -EMSGSIZE; 1190 1191 return 0; 1192 } 1193 1194 static int cmp_nla_table(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1195 { 1196 if (a->table != b->table) 1197 return 1; 1198 1199 return 0; 1200 } 1201 1202 static struct 1203 seg6_end_dt_info *seg6_possible_end_dt_info(struct seg6_local_lwt *slwt) 1204 { 1205 #ifdef CONFIG_NET_L3_MASTER_DEV 1206 return &slwt->dt_info; 1207 #else 1208 return ERR_PTR(-EOPNOTSUPP); 1209 #endif 1210 } 1211 1212 static int parse_nla_vrftable(struct nlattr **attrs, 1213 struct seg6_local_lwt *slwt) 1214 { 1215 struct seg6_end_dt_info *info = seg6_possible_end_dt_info(slwt); 1216 1217 if (IS_ERR(info)) 1218 return PTR_ERR(info); 1219 1220 info->vrf_table = nla_get_u32(attrs[SEG6_LOCAL_VRFTABLE]); 1221 1222 return 0; 1223 } 1224 1225 static int put_nla_vrftable(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1226 { 1227 struct seg6_end_dt_info *info = seg6_possible_end_dt_info(slwt); 1228 1229 if (IS_ERR(info)) 1230 return PTR_ERR(info); 1231 1232 if (nla_put_u32(skb, SEG6_LOCAL_VRFTABLE, info->vrf_table)) 1233 return -EMSGSIZE; 1234 1235 return 0; 1236 } 1237 1238 static int cmp_nla_vrftable(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1239 { 1240 struct seg6_end_dt_info *info_a = seg6_possible_end_dt_info(a); 1241 struct seg6_end_dt_info *info_b = seg6_possible_end_dt_info(b); 1242 1243 if (info_a->vrf_table != info_b->vrf_table) 1244 return 1; 1245 1246 return 0; 1247 } 1248 1249 static int parse_nla_nh4(struct nlattr **attrs, struct seg6_local_lwt *slwt) 1250 { 1251 memcpy(&slwt->nh4, nla_data(attrs[SEG6_LOCAL_NH4]), 1252 sizeof(struct in_addr)); 1253 1254 return 0; 1255 } 1256 1257 static int put_nla_nh4(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1258 { 1259 struct nlattr *nla; 1260 1261 nla = nla_reserve(skb, SEG6_LOCAL_NH4, sizeof(struct in_addr)); 1262 if (!nla) 1263 return -EMSGSIZE; 1264 1265 memcpy(nla_data(nla), &slwt->nh4, sizeof(struct in_addr)); 1266 1267 return 0; 1268 } 1269 1270 static int cmp_nla_nh4(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1271 { 1272 return memcmp(&a->nh4, &b->nh4, sizeof(struct in_addr)); 1273 } 1274 1275 static int parse_nla_nh6(struct nlattr **attrs, struct seg6_local_lwt *slwt) 1276 { 1277 memcpy(&slwt->nh6, nla_data(attrs[SEG6_LOCAL_NH6]), 1278 sizeof(struct in6_addr)); 1279 1280 return 0; 1281 } 1282 1283 static int put_nla_nh6(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1284 { 1285 struct nlattr *nla; 1286 1287 nla = nla_reserve(skb, SEG6_LOCAL_NH6, sizeof(struct in6_addr)); 1288 if (!nla) 1289 return -EMSGSIZE; 1290 1291 memcpy(nla_data(nla), &slwt->nh6, sizeof(struct in6_addr)); 1292 1293 return 0; 1294 } 1295 1296 static int cmp_nla_nh6(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1297 { 1298 return memcmp(&a->nh6, &b->nh6, sizeof(struct in6_addr)); 1299 } 1300 1301 static int parse_nla_iif(struct nlattr **attrs, struct seg6_local_lwt *slwt) 1302 { 1303 slwt->iif = nla_get_u32(attrs[SEG6_LOCAL_IIF]); 1304 1305 return 0; 1306 } 1307 1308 static int put_nla_iif(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1309 { 1310 if (nla_put_u32(skb, SEG6_LOCAL_IIF, slwt->iif)) 1311 return -EMSGSIZE; 1312 1313 return 0; 1314 } 1315 1316 static int cmp_nla_iif(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1317 { 1318 if (a->iif != b->iif) 1319 return 1; 1320 1321 return 0; 1322 } 1323 1324 static int parse_nla_oif(struct nlattr **attrs, struct seg6_local_lwt *slwt) 1325 { 1326 slwt->oif = nla_get_u32(attrs[SEG6_LOCAL_OIF]); 1327 1328 return 0; 1329 } 1330 1331 static int put_nla_oif(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1332 { 1333 if (nla_put_u32(skb, SEG6_LOCAL_OIF, slwt->oif)) 1334 return -EMSGSIZE; 1335 1336 return 0; 1337 } 1338 1339 static int cmp_nla_oif(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1340 { 1341 if (a->oif != b->oif) 1342 return 1; 1343 1344 return 0; 1345 } 1346 1347 #define MAX_PROG_NAME 256 1348 static const struct nla_policy bpf_prog_policy[SEG6_LOCAL_BPF_PROG_MAX + 1] = { 1349 [SEG6_LOCAL_BPF_PROG] = { .type = NLA_U32, }, 1350 [SEG6_LOCAL_BPF_PROG_NAME] = { .type = NLA_NUL_STRING, 1351 .len = MAX_PROG_NAME }, 1352 }; 1353 1354 static int parse_nla_bpf(struct nlattr **attrs, struct seg6_local_lwt *slwt) 1355 { 1356 struct nlattr *tb[SEG6_LOCAL_BPF_PROG_MAX + 1]; 1357 struct bpf_prog *p; 1358 int ret; 1359 u32 fd; 1360 1361 ret = nla_parse_nested_deprecated(tb, SEG6_LOCAL_BPF_PROG_MAX, 1362 attrs[SEG6_LOCAL_BPF], 1363 bpf_prog_policy, NULL); 1364 if (ret < 0) 1365 return ret; 1366 1367 if (!tb[SEG6_LOCAL_BPF_PROG] || !tb[SEG6_LOCAL_BPF_PROG_NAME]) 1368 return -EINVAL; 1369 1370 slwt->bpf.name = nla_memdup(tb[SEG6_LOCAL_BPF_PROG_NAME], GFP_KERNEL); 1371 if (!slwt->bpf.name) 1372 return -ENOMEM; 1373 1374 fd = nla_get_u32(tb[SEG6_LOCAL_BPF_PROG]); 1375 p = bpf_prog_get_type(fd, BPF_PROG_TYPE_LWT_SEG6LOCAL); 1376 if (IS_ERR(p)) { 1377 kfree(slwt->bpf.name); 1378 return PTR_ERR(p); 1379 } 1380 1381 slwt->bpf.prog = p; 1382 return 0; 1383 } 1384 1385 static int put_nla_bpf(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1386 { 1387 struct nlattr *nest; 1388 1389 if (!slwt->bpf.prog) 1390 return 0; 1391 1392 nest = nla_nest_start_noflag(skb, SEG6_LOCAL_BPF); 1393 if (!nest) 1394 return -EMSGSIZE; 1395 1396 if (nla_put_u32(skb, SEG6_LOCAL_BPF_PROG, slwt->bpf.prog->aux->id)) 1397 return -EMSGSIZE; 1398 1399 if (slwt->bpf.name && 1400 nla_put_string(skb, SEG6_LOCAL_BPF_PROG_NAME, slwt->bpf.name)) 1401 return -EMSGSIZE; 1402 1403 return nla_nest_end(skb, nest); 1404 } 1405 1406 static int cmp_nla_bpf(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1407 { 1408 if (!a->bpf.name && !b->bpf.name) 1409 return 0; 1410 1411 if (!a->bpf.name || !b->bpf.name) 1412 return 1; 1413 1414 return strcmp(a->bpf.name, b->bpf.name); 1415 } 1416 1417 static void destroy_attr_bpf(struct seg6_local_lwt *slwt) 1418 { 1419 kfree(slwt->bpf.name); 1420 if (slwt->bpf.prog) 1421 bpf_prog_put(slwt->bpf.prog); 1422 } 1423 1424 static const struct 1425 nla_policy seg6_local_counters_policy[SEG6_LOCAL_CNT_MAX + 1] = { 1426 [SEG6_LOCAL_CNT_PACKETS] = { .type = NLA_U64 }, 1427 [SEG6_LOCAL_CNT_BYTES] = { .type = NLA_U64 }, 1428 [SEG6_LOCAL_CNT_ERRORS] = { .type = NLA_U64 }, 1429 }; 1430 1431 static int parse_nla_counters(struct nlattr **attrs, 1432 struct seg6_local_lwt *slwt) 1433 { 1434 struct pcpu_seg6_local_counters __percpu *pcounters; 1435 struct nlattr *tb[SEG6_LOCAL_CNT_MAX + 1]; 1436 int ret; 1437 1438 ret = nla_parse_nested_deprecated(tb, SEG6_LOCAL_CNT_MAX, 1439 attrs[SEG6_LOCAL_COUNTERS], 1440 seg6_local_counters_policy, NULL); 1441 if (ret < 0) 1442 return ret; 1443 1444 /* basic support for SRv6 Behavior counters requires at least: 1445 * packets, bytes and errors. 1446 */ 1447 if (!tb[SEG6_LOCAL_CNT_PACKETS] || !tb[SEG6_LOCAL_CNT_BYTES] || 1448 !tb[SEG6_LOCAL_CNT_ERRORS]) 1449 return -EINVAL; 1450 1451 /* counters are always zero initialized */ 1452 pcounters = seg6_local_alloc_pcpu_counters(GFP_KERNEL); 1453 if (!pcounters) 1454 return -ENOMEM; 1455 1456 slwt->pcpu_counters = pcounters; 1457 1458 return 0; 1459 } 1460 1461 static int seg6_local_fill_nla_counters(struct sk_buff *skb, 1462 struct seg6_local_counters *counters) 1463 { 1464 if (nla_put_u64_64bit(skb, SEG6_LOCAL_CNT_PACKETS, counters->packets, 1465 SEG6_LOCAL_CNT_PAD)) 1466 return -EMSGSIZE; 1467 1468 if (nla_put_u64_64bit(skb, SEG6_LOCAL_CNT_BYTES, counters->bytes, 1469 SEG6_LOCAL_CNT_PAD)) 1470 return -EMSGSIZE; 1471 1472 if (nla_put_u64_64bit(skb, SEG6_LOCAL_CNT_ERRORS, counters->errors, 1473 SEG6_LOCAL_CNT_PAD)) 1474 return -EMSGSIZE; 1475 1476 return 0; 1477 } 1478 1479 static int put_nla_counters(struct sk_buff *skb, struct seg6_local_lwt *slwt) 1480 { 1481 struct seg6_local_counters counters = { 0, 0, 0 }; 1482 struct nlattr *nest; 1483 int rc, i; 1484 1485 nest = nla_nest_start(skb, SEG6_LOCAL_COUNTERS); 1486 if (!nest) 1487 return -EMSGSIZE; 1488 1489 for_each_possible_cpu(i) { 1490 struct pcpu_seg6_local_counters *pcounters; 1491 u64 packets, bytes, errors; 1492 unsigned int start; 1493 1494 pcounters = per_cpu_ptr(slwt->pcpu_counters, i); 1495 do { 1496 start = u64_stats_fetch_begin_irq(&pcounters->syncp); 1497 1498 packets = u64_stats_read(&pcounters->packets); 1499 bytes = u64_stats_read(&pcounters->bytes); 1500 errors = u64_stats_read(&pcounters->errors); 1501 1502 } while (u64_stats_fetch_retry_irq(&pcounters->syncp, start)); 1503 1504 counters.packets += packets; 1505 counters.bytes += bytes; 1506 counters.errors += errors; 1507 } 1508 1509 rc = seg6_local_fill_nla_counters(skb, &counters); 1510 if (rc < 0) { 1511 nla_nest_cancel(skb, nest); 1512 return rc; 1513 } 1514 1515 return nla_nest_end(skb, nest); 1516 } 1517 1518 static int cmp_nla_counters(struct seg6_local_lwt *a, struct seg6_local_lwt *b) 1519 { 1520 /* a and b are equal if both have pcpu_counters set or not */ 1521 return (!!((unsigned long)a->pcpu_counters)) ^ 1522 (!!((unsigned long)b->pcpu_counters)); 1523 } 1524 1525 static void destroy_attr_counters(struct seg6_local_lwt *slwt) 1526 { 1527 free_percpu(slwt->pcpu_counters); 1528 } 1529 1530 struct seg6_action_param { 1531 int (*parse)(struct nlattr **attrs, struct seg6_local_lwt *slwt); 1532 int (*put)(struct sk_buff *skb, struct seg6_local_lwt *slwt); 1533 int (*cmp)(struct seg6_local_lwt *a, struct seg6_local_lwt *b); 1534 1535 /* optional destroy() callback useful for releasing resources which 1536 * have been previously acquired in the corresponding parse() 1537 * function. 1538 */ 1539 void (*destroy)(struct seg6_local_lwt *slwt); 1540 }; 1541 1542 static struct seg6_action_param seg6_action_params[SEG6_LOCAL_MAX + 1] = { 1543 [SEG6_LOCAL_SRH] = { .parse = parse_nla_srh, 1544 .put = put_nla_srh, 1545 .cmp = cmp_nla_srh, 1546 .destroy = destroy_attr_srh }, 1547 1548 [SEG6_LOCAL_TABLE] = { .parse = parse_nla_table, 1549 .put = put_nla_table, 1550 .cmp = cmp_nla_table }, 1551 1552 [SEG6_LOCAL_NH4] = { .parse = parse_nla_nh4, 1553 .put = put_nla_nh4, 1554 .cmp = cmp_nla_nh4 }, 1555 1556 [SEG6_LOCAL_NH6] = { .parse = parse_nla_nh6, 1557 .put = put_nla_nh6, 1558 .cmp = cmp_nla_nh6 }, 1559 1560 [SEG6_LOCAL_IIF] = { .parse = parse_nla_iif, 1561 .put = put_nla_iif, 1562 .cmp = cmp_nla_iif }, 1563 1564 [SEG6_LOCAL_OIF] = { .parse = parse_nla_oif, 1565 .put = put_nla_oif, 1566 .cmp = cmp_nla_oif }, 1567 1568 [SEG6_LOCAL_BPF] = { .parse = parse_nla_bpf, 1569 .put = put_nla_bpf, 1570 .cmp = cmp_nla_bpf, 1571 .destroy = destroy_attr_bpf }, 1572 1573 [SEG6_LOCAL_VRFTABLE] = { .parse = parse_nla_vrftable, 1574 .put = put_nla_vrftable, 1575 .cmp = cmp_nla_vrftable }, 1576 1577 [SEG6_LOCAL_COUNTERS] = { .parse = parse_nla_counters, 1578 .put = put_nla_counters, 1579 .cmp = cmp_nla_counters, 1580 .destroy = destroy_attr_counters }, 1581 }; 1582 1583 /* call the destroy() callback (if available) for each set attribute in 1584 * @parsed_attrs, starting from the first attribute up to the @max_parsed 1585 * (excluded) attribute. 1586 */ 1587 static void __destroy_attrs(unsigned long parsed_attrs, int max_parsed, 1588 struct seg6_local_lwt *slwt) 1589 { 1590 struct seg6_action_param *param; 1591 int i; 1592 1593 /* Every required seg6local attribute is identified by an ID which is 1594 * encoded as a flag (i.e: 1 << ID) in the 'attrs' bitmask; 1595 * 1596 * We scan the 'parsed_attrs' bitmask, starting from the first attribute 1597 * up to the @max_parsed (excluded) attribute. 1598 * For each set attribute, we retrieve the corresponding destroy() 1599 * callback. If the callback is not available, then we skip to the next 1600 * attribute; otherwise, we call the destroy() callback. 1601 */ 1602 for (i = 0; i < max_parsed; ++i) { 1603 if (!(parsed_attrs & SEG6_F_ATTR(i))) 1604 continue; 1605 1606 param = &seg6_action_params[i]; 1607 1608 if (param->destroy) 1609 param->destroy(slwt); 1610 } 1611 } 1612 1613 /* release all the resources that may have been acquired during parsing 1614 * operations. 1615 */ 1616 static void destroy_attrs(struct seg6_local_lwt *slwt) 1617 { 1618 unsigned long attrs = slwt->desc->attrs | slwt->parsed_optattrs; 1619 1620 __destroy_attrs(attrs, SEG6_LOCAL_MAX + 1, slwt); 1621 } 1622 1623 static int parse_nla_optional_attrs(struct nlattr **attrs, 1624 struct seg6_local_lwt *slwt) 1625 { 1626 struct seg6_action_desc *desc = slwt->desc; 1627 unsigned long parsed_optattrs = 0; 1628 struct seg6_action_param *param; 1629 int err, i; 1630 1631 for (i = 0; i < SEG6_LOCAL_MAX + 1; ++i) { 1632 if (!(desc->optattrs & SEG6_F_ATTR(i)) || !attrs[i]) 1633 continue; 1634 1635 /* once here, the i-th attribute is provided by the 1636 * userspace AND it is identified optional as well. 1637 */ 1638 param = &seg6_action_params[i]; 1639 1640 err = param->parse(attrs, slwt); 1641 if (err < 0) 1642 goto parse_optattrs_err; 1643 1644 /* current attribute has been correctly parsed */ 1645 parsed_optattrs |= SEG6_F_ATTR(i); 1646 } 1647 1648 /* store in the tunnel state all the optional attributed successfully 1649 * parsed. 1650 */ 1651 slwt->parsed_optattrs = parsed_optattrs; 1652 1653 return 0; 1654 1655 parse_optattrs_err: 1656 __destroy_attrs(parsed_optattrs, i, slwt); 1657 1658 return err; 1659 } 1660 1661 /* call the custom constructor of the behavior during its initialization phase 1662 * and after that all its attributes have been parsed successfully. 1663 */ 1664 static int 1665 seg6_local_lwtunnel_build_state(struct seg6_local_lwt *slwt, const void *cfg, 1666 struct netlink_ext_ack *extack) 1667 { 1668 struct seg6_action_desc *desc = slwt->desc; 1669 struct seg6_local_lwtunnel_ops *ops; 1670 1671 ops = &desc->slwt_ops; 1672 if (!ops->build_state) 1673 return 0; 1674 1675 return ops->build_state(slwt, cfg, extack); 1676 } 1677 1678 /* call the custom destructor of the behavior which is invoked before the 1679 * tunnel is going to be destroyed. 1680 */ 1681 static void seg6_local_lwtunnel_destroy_state(struct seg6_local_lwt *slwt) 1682 { 1683 struct seg6_action_desc *desc = slwt->desc; 1684 struct seg6_local_lwtunnel_ops *ops; 1685 1686 ops = &desc->slwt_ops; 1687 if (!ops->destroy_state) 1688 return; 1689 1690 ops->destroy_state(slwt); 1691 } 1692 1693 static int parse_nla_action(struct nlattr **attrs, struct seg6_local_lwt *slwt) 1694 { 1695 struct seg6_action_param *param; 1696 struct seg6_action_desc *desc; 1697 unsigned long invalid_attrs; 1698 int i, err; 1699 1700 desc = __get_action_desc(slwt->action); 1701 if (!desc) 1702 return -EINVAL; 1703 1704 if (!desc->input) 1705 return -EOPNOTSUPP; 1706 1707 slwt->desc = desc; 1708 slwt->headroom += desc->static_headroom; 1709 1710 /* Forcing the desc->optattrs *set* and the desc->attrs *set* to be 1711 * disjoined, this allow us to release acquired resources by optional 1712 * attributes and by required attributes independently from each other 1713 * without any interference. 1714 * In other terms, we are sure that we do not release some the acquired 1715 * resources twice. 1716 * 1717 * Note that if an attribute is configured both as required and as 1718 * optional, it means that the user has messed something up in the 1719 * seg6_action_table. Therefore, this check is required for SRv6 1720 * behaviors to work properly. 1721 */ 1722 invalid_attrs = desc->attrs & desc->optattrs; 1723 if (invalid_attrs) { 1724 WARN_ONCE(1, 1725 "An attribute cannot be both required AND optional"); 1726 return -EINVAL; 1727 } 1728 1729 /* parse the required attributes */ 1730 for (i = 0; i < SEG6_LOCAL_MAX + 1; i++) { 1731 if (desc->attrs & SEG6_F_ATTR(i)) { 1732 if (!attrs[i]) 1733 return -EINVAL; 1734 1735 param = &seg6_action_params[i]; 1736 1737 err = param->parse(attrs, slwt); 1738 if (err < 0) 1739 goto parse_attrs_err; 1740 } 1741 } 1742 1743 /* parse the optional attributes, if any */ 1744 err = parse_nla_optional_attrs(attrs, slwt); 1745 if (err < 0) 1746 goto parse_attrs_err; 1747 1748 return 0; 1749 1750 parse_attrs_err: 1751 /* release any resource that may have been acquired during the i-1 1752 * parse() operations. 1753 */ 1754 __destroy_attrs(desc->attrs, i, slwt); 1755 1756 return err; 1757 } 1758 1759 static int seg6_local_build_state(struct net *net, struct nlattr *nla, 1760 unsigned int family, const void *cfg, 1761 struct lwtunnel_state **ts, 1762 struct netlink_ext_ack *extack) 1763 { 1764 struct nlattr *tb[SEG6_LOCAL_MAX + 1]; 1765 struct lwtunnel_state *newts; 1766 struct seg6_local_lwt *slwt; 1767 int err; 1768 1769 if (family != AF_INET6) 1770 return -EINVAL; 1771 1772 err = nla_parse_nested_deprecated(tb, SEG6_LOCAL_MAX, nla, 1773 seg6_local_policy, extack); 1774 1775 if (err < 0) 1776 return err; 1777 1778 if (!tb[SEG6_LOCAL_ACTION]) 1779 return -EINVAL; 1780 1781 newts = lwtunnel_state_alloc(sizeof(*slwt)); 1782 if (!newts) 1783 return -ENOMEM; 1784 1785 slwt = seg6_local_lwtunnel(newts); 1786 slwt->action = nla_get_u32(tb[SEG6_LOCAL_ACTION]); 1787 1788 err = parse_nla_action(tb, slwt); 1789 if (err < 0) 1790 goto out_free; 1791 1792 err = seg6_local_lwtunnel_build_state(slwt, cfg, extack); 1793 if (err < 0) 1794 goto out_destroy_attrs; 1795 1796 newts->type = LWTUNNEL_ENCAP_SEG6_LOCAL; 1797 newts->flags = LWTUNNEL_STATE_INPUT_REDIRECT; 1798 newts->headroom = slwt->headroom; 1799 1800 *ts = newts; 1801 1802 return 0; 1803 1804 out_destroy_attrs: 1805 destroy_attrs(slwt); 1806 out_free: 1807 kfree(newts); 1808 return err; 1809 } 1810 1811 static void seg6_local_destroy_state(struct lwtunnel_state *lwt) 1812 { 1813 struct seg6_local_lwt *slwt = seg6_local_lwtunnel(lwt); 1814 1815 seg6_local_lwtunnel_destroy_state(slwt); 1816 1817 destroy_attrs(slwt); 1818 1819 return; 1820 } 1821 1822 static int seg6_local_fill_encap(struct sk_buff *skb, 1823 struct lwtunnel_state *lwt) 1824 { 1825 struct seg6_local_lwt *slwt = seg6_local_lwtunnel(lwt); 1826 struct seg6_action_param *param; 1827 unsigned long attrs; 1828 int i, err; 1829 1830 if (nla_put_u32(skb, SEG6_LOCAL_ACTION, slwt->action)) 1831 return -EMSGSIZE; 1832 1833 attrs = slwt->desc->attrs | slwt->parsed_optattrs; 1834 1835 for (i = 0; i < SEG6_LOCAL_MAX + 1; i++) { 1836 if (attrs & SEG6_F_ATTR(i)) { 1837 param = &seg6_action_params[i]; 1838 err = param->put(skb, slwt); 1839 if (err < 0) 1840 return err; 1841 } 1842 } 1843 1844 return 0; 1845 } 1846 1847 static int seg6_local_get_encap_size(struct lwtunnel_state *lwt) 1848 { 1849 struct seg6_local_lwt *slwt = seg6_local_lwtunnel(lwt); 1850 unsigned long attrs; 1851 int nlsize; 1852 1853 nlsize = nla_total_size(4); /* action */ 1854 1855 attrs = slwt->desc->attrs | slwt->parsed_optattrs; 1856 1857 if (attrs & SEG6_F_ATTR(SEG6_LOCAL_SRH)) 1858 nlsize += nla_total_size((slwt->srh->hdrlen + 1) << 3); 1859 1860 if (attrs & SEG6_F_ATTR(SEG6_LOCAL_TABLE)) 1861 nlsize += nla_total_size(4); 1862 1863 if (attrs & SEG6_F_ATTR(SEG6_LOCAL_NH4)) 1864 nlsize += nla_total_size(4); 1865 1866 if (attrs & SEG6_F_ATTR(SEG6_LOCAL_NH6)) 1867 nlsize += nla_total_size(16); 1868 1869 if (attrs & SEG6_F_ATTR(SEG6_LOCAL_IIF)) 1870 nlsize += nla_total_size(4); 1871 1872 if (attrs & SEG6_F_ATTR(SEG6_LOCAL_OIF)) 1873 nlsize += nla_total_size(4); 1874 1875 if (attrs & SEG6_F_ATTR(SEG6_LOCAL_BPF)) 1876 nlsize += nla_total_size(sizeof(struct nlattr)) + 1877 nla_total_size(MAX_PROG_NAME) + 1878 nla_total_size(4); 1879 1880 if (attrs & SEG6_F_ATTR(SEG6_LOCAL_VRFTABLE)) 1881 nlsize += nla_total_size(4); 1882 1883 if (attrs & SEG6_F_LOCAL_COUNTERS) 1884 nlsize += nla_total_size(0) + /* nest SEG6_LOCAL_COUNTERS */ 1885 /* SEG6_LOCAL_CNT_PACKETS */ 1886 nla_total_size_64bit(sizeof(__u64)) + 1887 /* SEG6_LOCAL_CNT_BYTES */ 1888 nla_total_size_64bit(sizeof(__u64)) + 1889 /* SEG6_LOCAL_CNT_ERRORS */ 1890 nla_total_size_64bit(sizeof(__u64)); 1891 1892 return nlsize; 1893 } 1894 1895 static int seg6_local_cmp_encap(struct lwtunnel_state *a, 1896 struct lwtunnel_state *b) 1897 { 1898 struct seg6_local_lwt *slwt_a, *slwt_b; 1899 struct seg6_action_param *param; 1900 unsigned long attrs_a, attrs_b; 1901 int i; 1902 1903 slwt_a = seg6_local_lwtunnel(a); 1904 slwt_b = seg6_local_lwtunnel(b); 1905 1906 if (slwt_a->action != slwt_b->action) 1907 return 1; 1908 1909 attrs_a = slwt_a->desc->attrs | slwt_a->parsed_optattrs; 1910 attrs_b = slwt_b->desc->attrs | slwt_b->parsed_optattrs; 1911 1912 if (attrs_a != attrs_b) 1913 return 1; 1914 1915 for (i = 0; i < SEG6_LOCAL_MAX + 1; i++) { 1916 if (attrs_a & SEG6_F_ATTR(i)) { 1917 param = &seg6_action_params[i]; 1918 if (param->cmp(slwt_a, slwt_b)) 1919 return 1; 1920 } 1921 } 1922 1923 return 0; 1924 } 1925 1926 static const struct lwtunnel_encap_ops seg6_local_ops = { 1927 .build_state = seg6_local_build_state, 1928 .destroy_state = seg6_local_destroy_state, 1929 .input = seg6_local_input, 1930 .fill_encap = seg6_local_fill_encap, 1931 .get_encap_size = seg6_local_get_encap_size, 1932 .cmp_encap = seg6_local_cmp_encap, 1933 .owner = THIS_MODULE, 1934 }; 1935 1936 int __init seg6_local_init(void) 1937 { 1938 /* If the max total number of defined attributes is reached, then your 1939 * kernel build stops here. 1940 * 1941 * This check is required to avoid arithmetic overflows when processing 1942 * behavior attributes and the maximum number of defined attributes 1943 * exceeds the allowed value. 1944 */ 1945 BUILD_BUG_ON(SEG6_LOCAL_MAX + 1 > BITS_PER_TYPE(unsigned long)); 1946 1947 return lwtunnel_encap_add_ops(&seg6_local_ops, 1948 LWTUNNEL_ENCAP_SEG6_LOCAL); 1949 } 1950 1951 void seg6_local_exit(void) 1952 { 1953 lwtunnel_encap_del_ops(&seg6_local_ops, LWTUNNEL_ENCAP_SEG6_LOCAL); 1954 } 1955