aboutsummaryrefslogtreecommitdiffstats
path: root/src/vnet/udp/udp.api
diff options
context:
space:
mode:
authorFlorin Coras <fcoras@cisco.com>2019-04-08 07:42:30 -0700
committerDave Barach <openvpp@barachs.net>2019-04-08 20:29:16 +0000
commitc5df8c71cc867d8120a25e4bd6d065aa63d9011c (patch)
tree3f5088d55054b2e971a4c1cfef79d4ed504f5099 /src/vnet/udp/udp.api
parentf3262a2dd0fd788b5ef3fc9592de17ef8bcbd897 (diff)
host stack: update stale copyright
Change-Id: I33cd6e44d126c73c1f4c16b2041ea607b4d7f39f Signed-off-by: Florin Coras <fcoras@cisco.com>
Diffstat (limited to 'src/vnet/udp/udp.api')
-rw-r--r--src/vnet/udp/udp.api2
1 files changed, 1 insertions, 1 deletions
diff --git a/src/vnet/udp/udp.api b/src/vnet/udp/udp.api
index 193b577ad3b..a15261cb9f5 100644
--- a/src/vnet/udp/udp.api
+++ b/src/vnet/udp/udp.api
@@ -1,6 +1,6 @@
/* Hey Emacs use -*- mode: C -*- */
/*
- * Copyright (c) 2018 Cisco and/or its affiliates.
+ * Copyright (c) 20182-2019 Cisco and/or its affiliates.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at:
182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953
/*
 * ethernet/arp.c: IP v4 ARP node
 *
 * Copyright (c) 2010 Cisco and/or its affiliates.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at:
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#include <vnet/arp/arp.h>
#include <vnet/arp/arp_packet.h>

#include <vnet/fib/ip4_fib.h>
#include <vnet/fib/fib_entry_src.h>
#include <vnet/adj/adj_nbr.h>
#include <vnet/adj/adj_mcast.h>
#include <vnet/pg/pg.h>

#include <vnet/ip-neighbor/ip_neighbor.h>
#include <vnet/ip-neighbor/ip4_neighbor.h>
#include <vnet/ip-neighbor/ip_neighbor_dp.h>

#include <vlibmemory/api.h>

/**
 * @file
 * @brief IPv4 ARP.
 *
 * This file contains code to manage the IPv4 ARP tables (IP Address
 * to MAC Address lookup).
 */

/**
 * @brief Per-interface ARP configuration and state
 */
typedef struct ethernet_arp_interface_t_
{
  /**
   * Is ARP enabled on this interface
   */
  u32 enabled;
} ethernet_arp_interface_t;

typedef struct
{
  /* Hash tables mapping name to opcode. */
  uword *opcode_by_name;

  /** Per interface state */
  ethernet_arp_interface_t *ethernet_arp_by_sw_if_index;

  /* ARP feature arc index */
  u8 feature_arc_index;
} ethernet_arp_main_t;

static ethernet_arp_main_t ethernet_arp_main;

static const u8 vrrp_prefix[] = { 0x00, 0x00, 0x5E, 0x00, 0x01 };

static uword
unformat_ethernet_arp_opcode_host_byte_order (unformat_input_t * input,
					      va_list * args)
{
  int *result = va_arg (*args, int *);
  ethernet_arp_main_t *am = &ethernet_arp_main;
  int x, i;

  /* Numeric opcode. */
  if (unformat (input, "0x%x", &x) || unformat (input, "%d", &x))
    {
      if (x >= (1 << 16))
	return 0;
      *result = x;
      return 1;
    }

  /* Named type. */
  if (unformat_user (input, unformat_vlib_number_by_name,
		     am->opcode_by_name, &i))
    {
      *result = i;
      return 1;
    }

  return 0;
}

static uword
unformat_ethernet_arp_opcode_net_byte_order (unformat_input_t * input,
					     va_list * args)
{
  int *result = va_arg (*args, int *);
  if (!unformat_user
      (input, unformat_ethernet_arp_opcode_host_byte_order, result))
    return 0;

  *result = clib_host_to_net_u16 ((u16) * result);
  return 1;
}

typedef struct
{
  u8 packet_data[64];
} ethernet_arp_input_trace_t;

static u8 *
format_ethernet_arp_input_trace (u8 * s, va_list * va)
{
  CLIB_UNUSED (vlib_main_t * vm) = va_arg (*va, vlib_main_t *);
  CLIB_UNUSED (vlib_node_t * node) = va_arg (*va, vlib_node_t *);
  ethernet_arp_input_trace_t *t = va_arg (*va, ethernet_arp_input_trace_t *);

  s = format (s, "%U",
	      format_ethernet_arp_header,
	      t->packet_data, sizeof (t->packet_data));

  return s;
}

static int
arp_is_enabled (ethernet_arp_main_t * am, u32 sw_if_index)
{
  if (vec_len (am->ethernet_arp_by_sw_if_index) <= sw_if_index)
    return 0;

  return (am->ethernet_arp_by_sw_if_index[sw_if_index].enabled);
}

static void
arp_enable (ethernet_arp_main_t * am, u32 sw_if_index)
{
  if (arp_is_enabled (am, sw_if_index))
    return;

  vec_validate (am->ethernet_arp_by_sw_if_index, sw_if_index);

  am->ethernet_arp_by_sw_if_index[sw_if_index].enabled = 1;

  vnet_feature_enable_disable ("arp", "arp-reply", sw_if_index, 1, NULL, 0);
  vnet_feature_enable_disable ("arp", "arp-disabled", sw_if_index, 0, NULL,
			       0);
}

static void
arp_disable (ethernet_arp_main_t * am, u32 sw_if_index)
{
  if (!arp_is_enabled (am, sw_if_index))
    return;

  vnet_feature_enable_disable ("arp", "arp-disabled", sw_if_index, 1, NULL,
			       0);
  vnet_feature_enable_disable ("arp", "arp-reply", sw_if_index, 0, NULL, 0);

  am->ethernet_arp_by_sw_if_index[sw_if_index].enabled = 0;
}

static int
arp_unnumbered (vlib_buffer_t * p0,
		u32 input_sw_if_index, u32 conn_sw_if_index)
{
  vnet_main_t *vnm = vnet_get_main ();
  vnet_interface_main_t *vim = &vnm->interface_main;
  vnet_sw_interface_t *si;

  /* verify that the input interface is unnumbered to the connected.
   * the connected interface is the interface on which the subnet is
   * configured */
  si = &vim->sw_interfaces[input_sw_if_index];

  if (!(si->flags & VNET_SW_INTERFACE_FLAG_UNNUMBERED &&
	(si->unnumbered_sw_if_index == conn_sw_if_index)))
    {
      /* the input interface is not unnumbered to the interface on which
       * the sub-net is configured that covers the ARP request.
       * So this is not the case for unnumbered.. */
      return 0;
    }

  return !0;
}

always_inline u32
arp_learn (u32 sw_if_index,
	   const ethernet_arp_ip4_over_ethernet_address_t * addr)
{
  /* *INDENT-OFF* */
  ip_neighbor_learn_t l = {
    .ip = {
      .ip.ip4 = addr->ip4,
      .version = AF_IP4,
    },
    .mac = addr->mac,
    .sw_if_index = sw_if_index,
  };
  /* *INDENT-ON* */

  ip_neighbor_learn_dp (&l);

  return (ARP_ERROR_L3_SRC_ADDRESS_LEARNED);
}

typedef enum arp_input_next_t_
{
  ARP_INPUT_NEXT_DROP,
  ARP_INPUT_NEXT_DISABLED,
  ARP_INPUT_N_NEXT,
} arp_input_next_t;

static uword
arp_input (vlib_main_t * vm, vlib_node_runtime_t * node, vlib_frame_t * frame)
{
  u32 n_left_from, next_index, *from, *to_next, n_left_to_next;
  ethernet_arp_main_t *am = &ethernet_arp_main;

  from = vlib_frame_vector_args (frame);
  n_left_from = frame->n_vectors;
  next_index = node->cached_next_index;

  if (node->flags & VLIB_NODE_FLAG_TRACE)
    vlib_trace_frame_buffers_only (vm, node, from, frame->n_vectors,
				   /* stride */ 1,
				   sizeof (ethernet_arp_input_trace_t));

  while (n_left_from > 0)
    {
      vlib_get_next_frame (vm, node, next_index, to_next, n_left_to_next);

      while (n_left_from > 0 && n_left_to_next > 0)
	{
	  const ethernet_arp_header_t *arp0;
	  arp_input_next_t next0;
	  vlib_buffer_t *p0;
	  u32 pi0, error0;

	  pi0 = to_next[0] = from[0];
	  from += 1;
	  to_next += 1;
	  n_left_from -= 1;
	  n_left_to_next -= 1;

	  p0 = vlib_get_buffer (vm, pi0);
	  arp0 = vlib_buffer_get_current (p0);

	  error0 = ARP_ERROR_REPLIES_SENT;
	  next0 = ARP_INPUT_NEXT_DROP;

	  error0 = (arp0->l2_type != clib_net_to_host_u16 (
				       ETHERNET_ARP_HARDWARE_TYPE_ethernet) ?
			    ARP_ERROR_L2_TYPE_NOT_ETHERNET :
			    error0);
	  error0 = (arp0->l3_type != clib_net_to_host_u16 (ETHERNET_TYPE_IP4) ?
			    ARP_ERROR_L3_TYPE_NOT_IP4 :
			    error0);
	  error0 = (0 == arp0->ip4_over_ethernet[0].ip4.as_u32 ?
			    ARP_ERROR_L3_DST_ADDRESS_UNSET :
			    error0);

	  if (ARP_ERROR_REPLIES_SENT == error0)
	    {
	      next0 = ARP_INPUT_NEXT_DISABLED;
	      vnet_feature_arc_start (am->feature_arc_index,
				      vnet_buffer (p0)->sw_if_index[VLIB_RX],
				      &next0, p0);
	    }
	  else
	    p0->error = node->errors[error0];

	  vlib_validate_buffer_enqueue_x1 (vm, node, next_index, to_next,
					   n_left_to_next, pi0, next0);
	}

      vlib_put_next_frame (vm, node, next_index, n_left_to_next);
    }

  return frame->n_vectors;
}

typedef enum arp_disabled_next_t_
{
  ARP_DISABLED_NEXT_DROP,
  ARP_DISABLED_N_NEXT,
} arp_disabled_next_t;

static uword
arp_disabled (vlib_main_t * vm,
	      vlib_node_runtime_t * node, vlib_frame_t * frame)
{
  u32 n_left_from, next_index, *from, *to_next, n_left_to_next;

  from = vlib_frame_vector_args (frame);
  n_left_from = frame->n_vectors;
  next_index = node->cached_next_index;

  if (node->flags & VLIB_NODE_FLAG_TRACE)
    vlib_trace_frame_buffers_only (vm, node, from, frame->n_vectors,
				   /* stride */ 1,
				   sizeof (ethernet_arp_input_trace_t));

  while (n_left_from > 0)
    {
      vlib_get_next_frame (vm, node, next_index, to_next, n_left_to_next);

      while (n_left_from > 0 && n_left_to_next > 0)
	{
	  arp_disabled_next_t next0 = ARP_DISABLED_NEXT_DROP;
	  vlib_buffer_t *p0;
	  u32 pi0, error0;

	  next0 = ARP_DISABLED_NEXT_DROP;
	  error0 = ARP_ERROR_DISABLED;

	  pi0 = to_next[0] = from[0];
	  from += 1;
	  to_next += 1;
	  n_left_from -= 1;
	  n_left_to_next -= 1;

	  p0 = vlib_get_buffer (vm, pi0);
	  p0->error = node->errors[error0];

	  vlib_validate_buffer_enqueue_x1 (vm, node, next_index, to_next,
					   n_left_to_next, pi0, next0);
	}

      vlib_put_next_frame (vm, node, next_index, n_left_to_next);
    }

  return frame->n_vectors;
}

enum arp_dst_fib_type
{
  ARP_DST_FIB_NONE,
  ARP_DST_FIB_ADJ,
  ARP_DST_FIB_CONN
};

/*
 * we're looking for FIB sources that indicate the destination
 * is attached. There may be interposed DPO prior to the one
 * we are looking for
 */
static enum arp_dst_fib_type
arp_dst_fib_check (const fib_node_index_t fei, fib_entry_flag_t * flags)
{
  const fib_entry_t *entry = fib_entry_get (fei);
  const fib_entry_src_t *entry_src;
  fib_source_t src;
  /* *INDENT-OFF* */
  FOR_EACH_SRC_ADDED(entry, entry_src, src,
  ({
    *flags = fib_entry_get_flags_for_source (fei, src);
    if (fib_entry_is_sourced (fei, FIB_SOURCE_ADJ))
        return ARP_DST_FIB_ADJ;
      else if (FIB_ENTRY_FLAG_CONNECTED & *flags)
        return ARP_DST_FIB_CONN;
  }))
  /* *INDENT-ON* */

  return ARP_DST_FIB_NONE;
}

static uword
arp_reply (vlib_main_t * vm, vlib_node_runtime_t * node, vlib_frame_t * frame)
{
  vnet_main_t *vnm = vnet_get_main ();
  u32 n_left_from, next_index, *from, *to_next;
  u32 n_replies_sent = 0;

  from = vlib_frame_vector_args (frame);
  n_left_from = frame->n_vectors;
  next_index = node->cached_next_index;

  if (node->flags & VLIB_NODE_FLAG_TRACE)
    vlib_trace_frame_buffers_only (vm, node, from, frame->n_vectors,
				   /* stride */ 1,
				   sizeof (ethernet_arp_input_trace_t));

  while (n_left_from > 0)
    {
      u32 n_left_to_next;

      vlib_get_next_frame (vm, node, next_index, to_next, n_left_to_next);

      while (n_left_from > 0 && n_left_to_next > 0)
	{
	  vlib_buffer_t *p0;
	  ethernet_arp_header_t *arp0;
	  ethernet_header_t *eth_rx;
	  const ip4_address_t *if_addr0;
	  u32 pi0, error0, next0, sw_if_index0, conn_sw_if_index0, fib_index0;
	  u8 dst_is_local0, is_vrrp_reply0;
	  fib_node_index_t dst_fei, src_fei;
	  const fib_prefix_t *pfx0;
	  fib_entry_flag_t src_flags, dst_flags;

	  pi0 = from[0];
	  to_next[0] = pi0;
	  from += 1;
	  to_next += 1;
	  n_left_from -= 1;
	  n_left_to_next -= 1;

	  p0 = vlib_get_buffer (vm, pi0);
	  arp0 = vlib_buffer_get_current (p0);
	  /* Fill in ethernet header. */
	  eth_rx = ethernet_buffer_get_header (p0);

	  next0 = ARP_REPLY_NEXT_DROP;
	  error0 = ARP_ERROR_REPLIES_SENT;
	  sw_if_index0 = vnet_buffer (p0)->sw_if_index[VLIB_RX];

	  /* Check that IP address is local and matches incoming interface. */
	  fib_index0 = ip4_fib_table_get_index_for_sw_if_index (sw_if_index0);
	  if (~0 == fib_index0)
	    {
	      error0 = ARP_ERROR_INTERFACE_NO_TABLE;
	      goto drop;

	    }

	  {
	    /*
	     * we're looking for FIB entries that indicate the source
	     * is attached. There may be more specific non-attached
	     * routes that match the source, but these do not influence
	     * whether we respond to an ARP request, i.e. they do not
	     * influence whether we are the correct way for the sender
	     * to reach us, they only affect how we reach the sender.
	     */
	    fib_entry_t *src_fib_entry;
	    const fib_prefix_t *pfx;
	    fib_entry_src_t *src;
	    fib_source_t source;
	    int attached;
	    int mask;

	    mask = 32;
	    attached = 0;

	    do
	      {
		src_fei = ip4_fib_table_lookup (ip4_fib_get (fib_index0),
						&arp0->
						ip4_over_ethernet[0].ip4,
						mask);
		src_fib_entry = fib_entry_get (src_fei);

		/*
		 * It's possible that the source that provides the
		 * flags we need, or the flags we must not have,
		 * is not the best source, so check then all.
		 */
                /* *INDENT-OFF* */
                FOR_EACH_SRC_ADDED(src_fib_entry, src, source,
                ({
                  src_flags = fib_entry_get_flags_for_source (src_fei, source);

                  /* Reject requests/replies with our local interface
                     address. */
                  if (FIB_ENTRY_FLAG_LOCAL & src_flags)
                    {
		      error0 = ARP_ERROR_L3_SRC_ADDRESS_IS_LOCAL;
		      /*
		       * When VPP has an interface whose address is also
		       * applied to a TAP interface on the host, then VPP's
		       * TAP interface will be unnumbered  to the 'real'
		       * interface and do proxy ARP from the host.
		       * The curious aspect of this setup is that ARP requests
		       * from the host will come from the VPP's own address.
		       * So don't drop immediately here, instead go see if this
		       * is a proxy ARP case.
		       */
		      goto next_feature;
		    }
		  /* A Source must also be local to subnet of matching
		   * interface address. */
		  if ((FIB_ENTRY_FLAG_ATTACHED & src_flags) ||
		      (FIB_ENTRY_FLAG_CONNECTED & src_flags))
		    {
		      attached = 1;
		      break;
		    }
		  /*
		   * else
		   *  The packet was sent from an address that is not
		   *  connected nor attached i.e. it is not from an
		   *  address that is covered by a link's sub-net,
		   *  nor is it a already learned host resp.
		   */
                }));
                /* *INDENT-ON* */

		/*
		 * shorter mask lookup for the next iteration.
		 */
		pfx = fib_entry_get_prefix (src_fei);
		mask = pfx->fp_len - 1;

		/*
		 * continue until we hit the default route or we find
		 * the attached we are looking for. The most likely
		 * outcome is we find the attached with the first source
		 * on the first lookup.
		 */
	      }
	    while (!attached &&
		   !fib_entry_is_sourced (src_fei, FIB_SOURCE_DEFAULT_ROUTE));

	    if (!attached)
	      {
		/*
		 * the matching route is a not attached, i.e. it was
		 * added as a result of routing, rather than interface/ARP
		 * configuration. If the matching route is not a host route
		 * (i.e. a /32)
		 */
		error0 = ARP_ERROR_L3_SRC_ADDRESS_NOT_LOCAL;
		goto drop;
	      }
	  }

	  dst_fei = ip4_fib_table_lookup (ip4_fib_get (fib_index0),
					  &arp0->ip4_over_ethernet[1].ip4,
					  32);
	  conn_sw_if_index0 = fib_entry_get_any_resolving_interface (dst_fei);

	  switch (arp_dst_fib_check (dst_fei, &dst_flags))
	    {
	    case ARP_DST_FIB_ADJ:
	      /*
	       * We matched an adj-fib on ths source subnet (a /32 previously
	       * added as a result of ARP). If this request is a gratuitous
	       * ARP, then learn from it.
	       * The check for matching an adj-fib, is to prevent hosts
	       * from spamming us with gratuitous ARPS that might otherwise
	       * blow our ARP cache
	       */
	      if (conn_sw_if_index0 != sw_if_index0)
		error0 = ARP_ERROR_L3_DST_ADDRESS_NOT_LOCAL;
	      else if (arp0->ip4_over_ethernet[0].ip4.as_u32 ==
		       arp0->ip4_over_ethernet[1].ip4.as_u32)
		{
		  vlib_increment_simple_counter (
		    &ip_neighbor_counters[AF_IP4]
		       .ipnc[VLIB_RX][IP_NEIGHBOR_CTR_GRAT],
		    vm->thread_index, sw_if_index0, 1);
		  error0 =
		    arp_learn (sw_if_index0, &arp0->ip4_over_ethernet[0]);
		}
	      goto next_feature;
	    case ARP_DST_FIB_CONN:
	      /* destination is connected, continue to process */
	      break;
	    case ARP_DST_FIB_NONE:
	      /* destination is not connected, stop here */
	      error0 = ARP_ERROR_L3_DST_ADDRESS_NOT_LOCAL;
	      goto next_feature;
	    }

	  dst_is_local0 = (FIB_ENTRY_FLAG_LOCAL & dst_flags);
	  pfx0 = fib_entry_get_prefix (dst_fei);
	  if_addr0 = &pfx0->fp_addr.ip4;

	  is_vrrp_reply0 =
	    ((arp0->opcode ==
	      clib_host_to_net_u16 (ETHERNET_ARP_OPCODE_reply))
	     &&
	     (!memcmp
	      (arp0->ip4_over_ethernet[0].mac.bytes, vrrp_prefix,
	       sizeof (vrrp_prefix))));

	  /* Trash ARP packets whose ARP-level source addresses do not
	     match their L2-frame-level source addresses, unless it's
	     a reply from a VRRP virtual router */
	  if (!ethernet_mac_address_equal
	      (eth_rx->src_address,
	       arp0->ip4_over_ethernet[0].mac.bytes) && !is_vrrp_reply0)
	    {
	      error0 = ARP_ERROR_L2_ADDRESS_MISMATCH;
	      goto drop;
	    }

	  vlib_increment_simple_counter (
	    &ip_neighbor_counters[AF_IP4]
	       .ipnc[VLIB_RX][arp0->opcode == clib_host_to_net_u16 (
						ETHERNET_ARP_OPCODE_reply) ?
				      IP_NEIGHBOR_CTR_REPLY :
				      IP_NEIGHBOR_CTR_REQUEST],
	    vm->thread_index, sw_if_index0, 1);

	  /* Learn or update sender's mapping only for replies to addresses
	   * that are local to the subnet */
	  if (arp0->opcode ==
	      clib_host_to_net_u16 (ETHERNET_ARP_OPCODE_reply))
	    {
	      if (dst_is_local0)
		error0 =
		  arp_learn (sw_if_index0, &arp0->ip4_over_ethernet[0]);
	      else
		/* a reply for a non-local destination could be a GARP.
		 * GARPs for hosts we know were handled above, so this one
		 * we drop */
		error0 = ARP_ERROR_L3_DST_ADDRESS_NOT_LOCAL;

	      goto next_feature;
	    }
	  else if (arp0->opcode ==
		   clib_host_to_net_u16 (ETHERNET_ARP_OPCODE_request) &&
		   (dst_is_local0 == 0))
	    {
	      goto next_feature;
	    }

	  /* Honor unnumbered interface, if any */
	  if (sw_if_index0 != conn_sw_if_index0 ||
	      sw_if_index0 != fib_entry_get_resolving_interface (src_fei))
	    {
	      /*
	       * The interface the ARP is sent to or was received on is not the
	       * interface on which the covering prefix is configured.
	       * Maybe this is a case for unnumbered.
	       */
	      if (!arp_unnumbered (p0, sw_if_index0, conn_sw_if_index0))
		{
		  error0 = ARP_ERROR_UNNUMBERED_MISMATCH;
		  goto drop;
		}
	    }
	  if (arp0->ip4_over_ethernet[0].ip4.as_u32 ==
	      arp0->ip4_over_ethernet[1].ip4.as_u32)
	    {
	      error0 = ARP_ERROR_GRATUITOUS_ARP;
	      goto drop;
	    }

	  next0 = arp_mk_reply (vnm, p0, sw_if_index0,
				if_addr0, arp0, eth_rx);

	  /* We are going to reply to this request, so, in the absence of
	     errors, learn the sender */
	  if (!error0)
	    error0 = arp_learn (sw_if_index0, &arp0->ip4_over_ethernet[1]);

	  vlib_increment_simple_counter (
	    &ip_neighbor_counters[AF_IP4].ipnc[VLIB_TX][IP_NEIGHBOR_CTR_REPLY],
	    vm->thread_index, sw_if_index0, 1);
	  n_replies_sent += 1;
	  goto enqueue;

	next_feature:
	  vnet_feature_next (&next0, p0);

	drop:
	  p0->error = node->errors[error0];

	enqueue:
	  vlib_validate_buffer_enqueue_x1 (vm, node, next_index, to_next,
					   n_left_to_next, pi0, next0);
	}

      vlib_put_next_frame (vm, node, next_index, n_left_to_next);
    }

  vlib_error_count (vm, node->node_index, ARP_ERROR_REPLIES_SENT,
		    n_replies_sent);

  return frame->n_vectors;
}


/* *INDENT-OFF* */

VLIB_REGISTER_NODE (arp_input_node, static) =
{
  .function = arp_input,
  .name = "arp-input",
  .vector_size = sizeof (u32),
  .n_errors = ARP_N_ERROR,
  .error_counters = arp_error_counters,
  .n_next_nodes = ARP_INPUT_N_NEXT,
  .next_nodes = {
    [ARP_INPUT_NEXT_DROP] = "error-drop",
    [ARP_INPUT_NEXT_DISABLED] = "arp-disabled",
  },
  .format_buffer = format_ethernet_arp_header,
  .format_trace = format_ethernet_arp_input_trace,
};

VLIB_REGISTER_NODE (arp_disabled_node, static) =
{
  .function = arp_disabled,
  .name = "arp-disabled",
  .vector_size = sizeof (u32),
  .n_errors = ARP_N_ERROR,
  .error_counters = arp_error_counters,
  .n_next_nodes = ARP_DISABLED_N_NEXT,
  .next_nodes = {
    [ARP_INPUT_NEXT_DROP] = "error-drop",
  },
  .format_buffer = format_ethernet_arp_header,
  .format_trace = format_ethernet_arp_input_trace,
};

VLIB_REGISTER_NODE (arp_reply_node, static) =
{
  .function = arp_reply,
  .name = "arp-reply",
  .vector_size = sizeof (u32),
  .n_errors = ARP_N_ERROR,
  .error_counters = arp_error_counters,
  .n_next_nodes = ARP_REPLY_N_NEXT,
  .next_nodes = {
    [ARP_REPLY_NEXT_DROP] = "error-drop",
    [ARP_REPLY_NEXT_REPLY_TX] = "interface-output",
  },
  .format_buffer = format_ethernet_arp_header,
  .format_trace = format_ethernet_arp_input_trace,
};

/* Built-in ARP rx feature path definition */
VNET_FEATURE_ARC_INIT (arp_feat, static) =
{
  .arc_name = "arp",
  .start_nodes = VNET_FEATURES ("arp-input"),
  .last_in_arc = "error-drop",
  .arc_index_ptr = &ethernet_arp_main.feature_arc_index,
};

VNET_FEATURE_INIT (arp_reply_feat_node, static) =
{
  .arc_name = "arp",
  .node_name = "arp-reply",
  .runs_before = VNET_FEATURES ("arp-disabled"),
};

VNET_FEATURE_INIT (arp_proxy_feat_node, static) =
{
  .arc_name = "arp",
  .node_name = "arp-proxy",
  .runs_after = VNET_FEATURES ("arp-reply"),
  .runs_before = VNET_FEATURES ("arp-disabled"),
};

VNET_FEATURE_INIT (arp_disabled_feat_node, static) =
{
  .arc_name = "arp",
  .node_name = "arp-disabled",
  .runs_before = VNET_FEATURES ("error-drop"),
};

VNET_FEATURE_INIT (arp_drop_feat_node, static) =
{
  .arc_name = "arp",
  .node_name = "error-drop",
  .runs_before = 0,	/* last feature */
};

/* *INDENT-ON* */

typedef struct
{
  pg_edit_t l2_type, l3_type;
  pg_edit_t n_l2_address_bytes, n_l3_address_bytes;
  pg_edit_t opcode;
  struct
  {
    pg_edit_t mac;
    pg_edit_t ip4;
  } ip4_over_ethernet[2];
} pg_ethernet_arp_header_t;

static inline void
pg_ethernet_arp_header_init (pg_ethernet_arp_header_t * p)
{
  /* Initialize fields that are not bit fields in the IP header. */
#define _(f) pg_edit_init (&p->f, ethernet_arp_header_t, f);
  _(l2_type);
  _(l3_type);
  _(n_l2_address_bytes);
  _(n_l3_address_bytes);
  _(opcode);
  _(ip4_over_ethernet[0].mac);
  _(ip4_over_ethernet[0].ip4);
  _(ip4_over_ethernet[1].mac);
  _(ip4_over_ethernet[1].ip4);
#undef _
}

uword
unformat_pg_arp_header (unformat_input_t * input, va_list * args)
{
  pg_stream_t *s = va_arg (*args, pg_stream_t *);
  pg_ethernet_arp_header_t *p;
  u32 group_index;

  p = pg_create_edit_group (s, sizeof (p[0]), sizeof (ethernet_arp_header_t),
			    &group_index);
  pg_ethernet_arp_header_init (p);

  /* Defaults. */
  pg_edit_set_fixed (&p->l2_type, ETHERNET_ARP_HARDWARE_TYPE_ethernet);
  pg_edit_set_fixed (&p->l3_type, ETHERNET_TYPE_IP4);
  pg_edit_set_fixed (&p->n_l2_address_bytes, 6);
  pg_edit_set_fixed (&p->n_l3_address_bytes, 4);

  if (!unformat (input, "%U: %U/%U -> %U/%U",
		 unformat_pg_edit,
		 unformat_ethernet_arp_opcode_net_byte_order, &p->opcode,
		 unformat_pg_edit,
		 unformat_mac_address_t, &p->ip4_over_ethernet[0].mac,
		 unformat_pg_edit,
		 unformat_ip4_address, &p->ip4_over_ethernet[0].ip4,
		 unformat_pg_edit,
		 unformat_mac_address_t, &p->ip4_over_ethernet[1].mac,
		 unformat_pg_edit,
		 unformat_ip4_address, &p->ip4_over_ethernet[1].ip4))
    {
      /* Free up any edits we may have added. */
      pg_free_edit_group (s);
      return 0;
    }
  return 1;
}

/*
 * callback when an interface address is added or deleted
 */
static void
arp_enable_disable_interface (ip4_main_t * im,
			      uword opaque, u32 sw_if_index, u32 is_enable)
{
  ethernet_arp_main_t *am = &ethernet_arp_main;

  if (is_enable)
    arp_enable (am, sw_if_index);
  else
    arp_disable (am, sw_if_index);
}

/*
 * Remove any arp entries associated with the specified interface
 */
static clib_error_t *
vnet_arp_add_del_sw_interface (vnet_main_t * vnm, u32 sw_if_index, u32 is_add)
{
  ethernet_arp_main_t *am = &ethernet_arp_main;
  if (is_add)
    arp_disable (am, sw_if_index);
  return (NULL);
}

VNET_SW_INTERFACE_ADD_DEL_FUNCTION (vnet_arp_add_del_sw_interface);

const static ip_neighbor_vft_t arp_vft = {
  .inv_proxy4_add = arp_proxy_add,
  .inv_proxy4_del = arp_proxy_del,
  .inv_proxy4_enable = arp_proxy_enable,
  .inv_proxy4_disable = arp_proxy_disable,
};

static clib_error_t *
ethernet_arp_init (vlib_main_t * vm)
{
  ethernet_arp_main_t *am = &ethernet_arp_main;
  ip4_main_t *im = &ip4_main;
  pg_node_t *pn;

  ethernet_register_input_type (vm, ETHERNET_TYPE_ARP, arp_input_node.index);

  pn = pg_get_node (arp_input_node.index);
  pn->unformat_edit = unformat_pg_arp_header;

  am->opcode_by_name = hash_create_string (0, sizeof (uword));
#define _(o) hash_set_mem (am->opcode_by_name, #o, ETHERNET_ARP_OPCODE_##o);
  foreach_ethernet_arp_opcode;
#undef _

  /* don't trace ARP error packets */
  {
    vlib_node_runtime_t *rt =
      vlib_node_get_runtime (vm, arp_input_node.index);

    vnet_pcap_drop_trace_filter_add_del (rt->errors[ARP_ERROR_REPLIES_SENT],
					 1);
    vnet_pcap_drop_trace_filter_add_del (rt->errors[ARP_ERROR_DISABLED], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_L2_TYPE_NOT_ETHERNET], 1);
    vnet_pcap_drop_trace_filter_add_del (rt->errors[ARP_ERROR_L3_TYPE_NOT_IP4],
					 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_L3_SRC_ADDRESS_NOT_LOCAL], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_L3_DST_ADDRESS_NOT_LOCAL], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_L3_DST_ADDRESS_UNSET], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_L3_SRC_ADDRESS_IS_LOCAL], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_L3_SRC_ADDRESS_LEARNED], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_REPLIES_RECEIVED], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_OPCODE_NOT_REQUEST], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_PROXY_ARP_REPLIES_SENT], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_L2_ADDRESS_MISMATCH], 1);
    vnet_pcap_drop_trace_filter_add_del (rt->errors[ARP_ERROR_GRATUITOUS_ARP],
					 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_INTERFACE_NO_TABLE], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_INTERFACE_NOT_IP_ENABLED], 1);
    vnet_pcap_drop_trace_filter_add_del (
      rt->errors[ARP_ERROR_UNNUMBERED_MISMATCH], 1);
  }

  {
    ip4_enable_disable_interface_callback_t cb = {
      .function = arp_enable_disable_interface,
    };
    vec_add1 (im->enable_disable_interface_callbacks, cb);
  }

  ip_neighbor_register (AF_IP4, &arp_vft);

  return 0;
}

/* *INDENT-OFF* */
VLIB_INIT_FUNCTION (ethernet_arp_init) =
{
  .runs_after = VLIB_INITS("ethernet_init",
                           "ip_neighbor_init"),
};
/* *INDENT-ON* */

/*
 * fd.io coding-style-patch-verification: ON
 *
 * Local Variables:
 * eval: (c-set-style "gnu")
 * End:
 */