summaryrefslogtreecommitdiffstats
path: root/src
diff options
context:
space:
mode:
authorMohsin Kazmi <sykazmi@cisco.com>2018-09-27 15:00:32 +0200
committerNeale Ranns <nranns@cisco.com>2018-09-29 17:06:30 +0000
commit762d83c198c6c24cce4eb94a6d027f520d0b8383 (patch)
treea7e9c4f0cc872642a7e39cc6b2fcefa7d21070d1 /src
parent539663c401b087e4576e41af35144e73e1907c9d (diff)
vom: Fix the l2 port type in bridge domain
Change-Id: I8aba11142daeacce892ec529c2229ec54ee427e2 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Diffstat (limited to 'src')
-rw-r--r--src/vat/api_format.c9
1 files changed, 5 insertions, 4 deletions
diff --git a/src/vat/api_format.c b/src/vat/api_format.c
index 83b875e3be4..99ede1ea7fd 100644
--- a/src/vat/api_format.c
+++ b/src/vat/api_format.c
@@ -1389,12 +1389,13 @@ static void vl_api_bridge_domain_details_t_handler
u32 n_sw_ifs = ntohl (mp->n_sw_ifs);
int i;
- print (vam->ofp, "\n%-3s %-3s %-3s %-3s %-3s %-3s",
- " ID", "LRN", "FWD", "FLD", "BVI", "#IF");
+ print (vam->ofp, "\n%-3s %-3s %-3s %-3s %-3s %-6s %-3s",
+ " ID", "LRN", "FWD", "FLD", "BVI", "UU-FWD", "#IF");
- print (vam->ofp, "%3d %3d %3d %3d %3d %3d",
+ print (vam->ofp, "%3d %3d %3d %3d %3d %6d %3d",
ntohl (mp->bd_id), mp->learn, mp->forward,
- mp->flood, ntohl (mp->bvi_sw_if_index), n_sw_ifs);
+ mp->flood, ntohl (mp->bvi_sw_if_index),
+ ntohl (mp->uu_fwd_sw_if_index), n_sw_ifs);
if (n_sw_ifs)
{
174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005
/*
 * node.c - ipfix probe graph node
 *
 * Copyright (c) 2017 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 <vlib/vlib.h>
#include <vnet/vnet.h>
#include <vnet/pg/pg.h>
#include <vppinfra/crc32.h>
#include <vppinfra/error.h>
#include <flowprobe/flowprobe.h>
#include <vnet/ip/ip6_packet.h>

static void flowprobe_export_entry (vlib_main_t * vm, flowprobe_entry_t * e);

/**
 * @file flow record generator graph node
 */

typedef struct
{
  /** interface handle */
  u32 rx_sw_if_index;
  u32 tx_sw_if_index;
  /** packet timestamp */
  u64 timestamp;
  /** size of the buffer */
  u16 buffer_size;

  /** L2 information */
  u8 src_mac[6];
  u8 dst_mac[6];
  /** Ethertype */
  u16 ethertype;

  /** L3 information */
  ip46_address_t src_address;
  ip46_address_t dst_address;
  u8 protocol;
  u8 tos;

  /** L4 information */
  u16 src_port;
  u16 dst_port;

  flowprobe_variant_t which;
} flowprobe_trace_t;

static char *flowprobe_variant_strings[] = {
  [FLOW_VARIANT_IP4] = "IP4",
  [FLOW_VARIANT_IP6] = "IP6",
  [FLOW_VARIANT_L2] = "L2",
  [FLOW_VARIANT_L2_IP4] = "L2-IP4",
  [FLOW_VARIANT_L2_IP6] = "L2-IP6",
};

/* packet trace format function */
static u8 *
format_flowprobe_trace (u8 * s, va_list * args)
{
  CLIB_UNUSED (vlib_main_t * vm) = va_arg (*args, vlib_main_t *);
  CLIB_UNUSED (vlib_node_t * node) = va_arg (*args, vlib_node_t *);
  flowprobe_trace_t *t = va_arg (*args, flowprobe_trace_t *);
  uword indent = format_get_indent (s);

  s = format (s,
	      "FLOWPROBE[%s]: rx_sw_if_index %d, tx_sw_if_index %d, "
	      "timestamp %lld, size %d", flowprobe_variant_strings[t->which],
	      t->rx_sw_if_index, t->tx_sw_if_index,
	      t->timestamp, t->buffer_size);

  if (t->which == FLOW_VARIANT_L2)
    s = format (s, "\n%U -> %U", format_white_space, indent,
		format_ethernet_address, &t->src_mac,
		format_ethernet_address, &t->dst_mac);

  if (t->protocol > 0
      && (t->which == FLOW_VARIANT_L2_IP4 || t->which == FLOW_VARIANT_IP4
	  || t->which == FLOW_VARIANT_L2_IP6 || t->which == FLOW_VARIANT_IP6))
    s =
      format (s, "\n%U%U: %U -> %U", format_white_space, indent,
	      format_ip_protocol, t->protocol, format_ip46_address,
	      &t->src_address, IP46_TYPE_ANY, format_ip46_address,
	      &t->dst_address, IP46_TYPE_ANY);
  return s;
}

vlib_node_registration_t flowprobe_ip4_node;
vlib_node_registration_t flowprobe_ip6_node;
vlib_node_registration_t flowprobe_l2_node;

/* No counters at the moment */
#define foreach_flowprobe_error			\
_(COLLISION, "Hash table collisions")		\
_(BUFFER, "Buffer allocation error")		\
_(EXPORTED_PACKETS, "Exported packets")		\
_(INPATH, "Exported packets in path")

typedef enum
{
#define _(sym,str) FLOWPROBE_ERROR_##sym,
  foreach_flowprobe_error
#undef _
    FLOWPROBE_N_ERROR,
} flowprobe_error_t;

static char *flowprobe_error_strings[] = {
#define _(sym,string) string,
  foreach_flowprobe_error
#undef _
};

typedef enum
{
  FLOWPROBE_NEXT_DROP,
  FLOWPROBE_NEXT_IP4_LOOKUP,
  FLOWPROBE_N_NEXT,
} flowprobe_next_t;

#define FLOWPROBE_NEXT_NODES {					\
    [FLOWPROBE_NEXT_DROP] = "error-drop",			\
    [FLOWPROBE_NEXT_IP4_LOOKUP] = "ip4-lookup",		\
}

static inline flowprobe_variant_t
flowprobe_get_variant (flowprobe_variant_t which,
		       flowprobe_record_t flags, u16 ethertype)
{
  if (which == FLOW_VARIANT_L2
      && (flags & FLOW_RECORD_L3 || flags & FLOW_RECORD_L4))
    return ethertype == ETHERNET_TYPE_IP6 ? FLOW_VARIANT_L2_IP6 : ethertype ==
      ETHERNET_TYPE_IP4 ? FLOW_VARIANT_L2_IP4 : FLOW_VARIANT_L2;
  return which;
}

static inline u32
flowprobe_common_add (vlib_buffer_t * to_b, flowprobe_entry_t * e, u16 offset)
{
  u16 start = offset;

  /* Ingress interface */
  u32 rx_if = clib_host_to_net_u32 (e->key.rx_sw_if_index);
  clib_memcpy (to_b->data + offset, &rx_if, sizeof (rx_if));
  offset += sizeof (rx_if);

  /* Egress interface */
  u32 tx_if = clib_host_to_net_u32 (e->key.tx_sw_if_index);
  clib_memcpy (to_b->data + offset, &tx_if, sizeof (tx_if));
  offset += sizeof (tx_if);

  /* packet delta count */
  u64 packetdelta = clib_host_to_net_u64 (e->packetcount);
  clib_memcpy (to_b->data + offset, &packetdelta, sizeof (u64));
  offset += sizeof (u64);

  return offset - start;
}

static inline u32
flowprobe_l2_add (vlib_buffer_t * to_b, flowprobe_entry_t * e, u16 offset)
{
  u16 start = offset;

  /* src mac address */
  clib_memcpy (to_b->data + offset, &e->key.src_mac, 6);
  offset += 6;

  /* dst mac address */
  clib_memcpy (to_b->data + offset, &e->key.dst_mac, 6);
  offset += 6;

  /* ethertype */
  clib_memcpy (to_b->data + offset, &e->key.ethertype, 2);
  offset += 2;

  return offset - start;
}

static inline u32
flowprobe_l3_ip6_add (vlib_buffer_t * to_b, flowprobe_entry_t * e, u16 offset)
{
  u16 start = offset;

  /* ip6 src address */
  clib_memcpy (to_b->data + offset, &e->key.src_address,
	       sizeof (ip6_address_t));
  offset += sizeof (ip6_address_t);

  /* ip6 dst address */
  clib_memcpy (to_b->data + offset, &e->key.dst_address,
	       sizeof (ip6_address_t));
  offset += sizeof (ip6_address_t);

  /* Protocol */
  to_b->data[offset++] = e->key.protocol;

  /* octetDeltaCount */
  u64 octetdelta = clib_host_to_net_u64 (e->octetcount);
  clib_memcpy (to_b->data + offset, &octetdelta, sizeof (u64));
  offset += sizeof (u64);

  return offset - start;
}

static inline u32
flowprobe_l3_ip4_add (vlib_buffer_t * to_b, flowprobe_entry_t * e, u16 offset)
{
  u16 start = offset;

  /* ip4 src address */
  clib_memcpy (to_b->data + offset, &e->key.src_address.ip4,
	       sizeof (ip4_address_t));
  offset += sizeof (ip4_address_t);

  /* ip4 dst address */
  clib_memcpy (to_b->data + offset, &e->key.dst_address.ip4,
	       sizeof (ip4_address_t));
  offset += sizeof (ip4_address_t);

  /* Protocol */
  to_b->data[offset++] = e->key.protocol;

  /* octetDeltaCount */
  u64 octetdelta = clib_host_to_net_u64 (e->octetcount);
  clib_memcpy (to_b->data + offset, &octetdelta, sizeof (u64));
  offset += sizeof (u64);

  return offset - start;
}

static inline u32
flowprobe_l4_add (vlib_buffer_t * to_b, flowprobe_entry_t * e, u16 offset)
{
  u16 start = offset;

  /* src port */
  clib_memcpy (to_b->data + offset, &e->key.src_port, 2);
  offset += 2;

  /* dst port */
  clib_memcpy (to_b->data + offset, &e->key.dst_port, 2);
  offset += 2;

  return offset - start;
}

static inline u32
flowprobe_hash (flowprobe_key_t * k)
{
  flowprobe_main_t *fm = &flowprobe_main;
  u32 h = 0;

#ifdef clib_crc32c_uses_intrinsics
  h = clib_crc32c ((u8 *) k->as_u32, FLOWPROBE_KEY_IN_U32);
#else
  u64 tmp =
    k->as_u32[0] ^ k->as_u32[1] ^ k->as_u32[2] ^ k->as_u32[3] ^ k->as_u32[4];
  h = clib_xxhash (tmp);
#endif

  return h >> (32 - fm->ht_log2len);
}

flowprobe_entry_t *
flowprobe_lookup (u32 my_cpu_number, flowprobe_key_t * k, u32 * poolindex,
		  bool * collision)
{
  flowprobe_main_t *fm = &flowprobe_main;
  flowprobe_entry_t *e;
  u32 h;

  h = (fm->active_timer) ? flowprobe_hash (k) : 0;

  /* Lookup in the flow state pool */
  *poolindex = fm->hash_per_worker[my_cpu_number][h];
  if (*poolindex != ~0)
    {
      e = pool_elt_at_index (fm->pool_per_worker[my_cpu_number], *poolindex);
      if (e)
	{
	  /* Verify key or report collision */
	  if (memcmp (k, &e->key, sizeof (flowprobe_key_t)))
	    *collision = true;
	  return e;
	}
    }

  return 0;
}

flowprobe_entry_t *
flowprobe_create (u32 my_cpu_number, flowprobe_key_t * k, u32 * poolindex)
{
  flowprobe_main_t *fm = &flowprobe_main;
  u32 h;

  flowprobe_entry_t *e;

  /* Get my index */
  h = (fm->active_timer) ? flowprobe_hash (k) : 0;

  pool_get (fm->pool_per_worker[my_cpu_number], e);
  *poolindex = e - fm->pool_per_worker[my_cpu_number];
  fm->hash_per_worker[my_cpu_number][h] = *poolindex;

  e->key = *k;

  if (fm->passive_timer > 0)
    {
      e->passive_timer_handle = tw_timer_start_2t_1w_2048sl
	(fm->timers_per_worker[my_cpu_number], *poolindex, 0,
	 fm->passive_timer);
    }
  return e;
}

static inline void
add_to_flow_record_state (vlib_main_t * vm, vlib_node_runtime_t * node,
			  flowprobe_main_t * fm, vlib_buffer_t * b,
			  u64 timestamp, u16 length,
			  flowprobe_variant_t which, flowprobe_trace_t * t)
{
  if (fm->disabled)
    return;

  u32 my_cpu_number = vm->thread_index;
  u16 octets = 0;

  flowprobe_record_t flags = fm->context[which].flags;
  bool collect_ip4 = false, collect_ip6 = false;
  ASSERT (b);
  ethernet_header_t *eth = vlib_buffer_get_current (b);
  u16 ethertype = clib_net_to_host_u16 (eth->type);
  /* *INDENT-OFF* */
  flowprobe_key_t k = { {0} };
  /* *INDENT-ON* */
  ip4_header_t *ip4 = 0;
  ip6_header_t *ip6 = 0;
  udp_header_t *udp = 0;

  if (flags & FLOW_RECORD_L3 || flags & FLOW_RECORD_L4)
    {
      collect_ip4 = which == FLOW_VARIANT_L2_IP4 || which == FLOW_VARIANT_IP4;
      collect_ip6 = which == FLOW_VARIANT_L2_IP6 || which == FLOW_VARIANT_IP6;
    }

  k.rx_sw_if_index = vnet_buffer (b)->sw_if_index[VLIB_RX];
  k.tx_sw_if_index = vnet_buffer (b)->sw_if_index[VLIB_TX];

  k.which = which;

  if (flags & FLOW_RECORD_L2)
    {
      clib_memcpy (k.src_mac, eth->src_address, 6);
      clib_memcpy (k.dst_mac, eth->dst_address, 6);
      k.ethertype = ethertype;
    }
  if (collect_ip6 && ethertype == ETHERNET_TYPE_IP6)
    {
      ip6 = (ip6_header_t *) (eth + 1);
      udp = (udp_header_t *) (ip6 + 1);
      if (flags & FLOW_RECORD_L3)
	{
	  k.src_address.as_u64[0] = ip6->src_address.as_u64[0];
	  k.src_address.as_u64[1] = ip6->src_address.as_u64[1];
	  k.dst_address.as_u64[0] = ip6->dst_address.as_u64[0];
	  k.dst_address.as_u64[1] = ip6->dst_address.as_u64[1];
	}
      k.protocol = ip6->protocol;
      octets = clib_net_to_host_u16 (ip6->payload_length)
	+ sizeof (ip6_header_t);
    }
  if (collect_ip4 && ethertype == ETHERNET_TYPE_IP4)
    {
      ip4 = (ip4_header_t *) (eth + 1);
      udp = (udp_header_t *) (ip4 + 1);
      if (flags & FLOW_RECORD_L3)
	{
	  k.src_address.ip4.as_u32 = ip4->src_address.as_u32;
	  k.dst_address.ip4.as_u32 = ip4->dst_address.as_u32;
	}
      k.protocol = ip4->protocol;
      octets = clib_net_to_host_u16 (ip4->length);
    }
  if ((flags & FLOW_RECORD_L4) && udp &&
      (k.protocol == IP_PROTOCOL_TCP || k.protocol == IP_PROTOCOL_UDP))
    {
      k.src_port = udp->src_port;
      k.dst_port = udp->dst_port;
    }

  if (t)
    {
      t->rx_sw_if_index = k.rx_sw_if_index;
      t->tx_sw_if_index = k.tx_sw_if_index;
      clib_memcpy (t->src_mac, k.src_mac, 6);
      clib_memcpy (t->dst_mac, k.dst_mac, 6);
      t->ethertype = k.ethertype;
      t->src_address.ip4.as_u32 = k.src_address.ip4.as_u32;
      t->dst_address.ip4.as_u32 = k.dst_address.ip4.as_u32;
      t->protocol = k.protocol;
      t->src_port = k.src_port;
      t->dst_port = k.dst_port;
      t->which = k.which;
    }

  flowprobe_entry_t *e = 0;
  f64 now = vlib_time_now (vm);
  if (fm->active_timer > 0)
    {
      u32 poolindex = ~0;
      bool collision = false;

      e = flowprobe_lookup (my_cpu_number, &k, &poolindex, &collision);
      if (collision)
	{
	  /* Flush data and clean up entry for reuse. */
	  if (e->packetcount)
	    flowprobe_export_entry (vm, e);
	  e->key = k;
	  vlib_node_increment_counter (vm, node->node_index,
				       FLOWPROBE_ERROR_COLLISION, 1);
	}
      if (!e)			/* Create new entry */
	{
	  e = flowprobe_create (my_cpu_number, &k, &poolindex);
	  e->last_exported = now;
	}
    }
  else
    {
      e = &fm->stateless_entry[my_cpu_number];
      e->key = k;
    }

  if (e)
    {
      /* Updating entry */
      e->packetcount++;
      e->octetcount += octets;
      e->last_updated = now;

      if (fm->active_timer == 0
	  || (now > e->last_exported + fm->active_timer))
	flowprobe_export_entry (vm, e);
    }
}

static u16
flowprobe_get_headersize (void)
{
  return sizeof (ip4_header_t) + sizeof (udp_header_t) +
    sizeof (ipfix_message_header_t) + sizeof (ipfix_set_header_t);
}

static void
flowprobe_export_send (vlib_main_t * vm, vlib_buffer_t * b0,
		       flowprobe_variant_t which)
{
  flowprobe_main_t *fm = &flowprobe_main;
  flow_report_main_t *frm = &flow_report_main;
  vlib_frame_t *f;
  ip4_ipfix_template_packet_t *tp;
  ipfix_set_header_t *s;
  ipfix_message_header_t *h;
  ip4_header_t *ip;
  udp_header_t *udp;
  flowprobe_record_t flags = fm->context[which].flags;
  u32 my_cpu_number = vm->thread_index;

  /* Fill in header */
  flow_report_stream_t *stream;

  /* Nothing to send */
  if (fm->context[which].next_record_offset_per_worker[my_cpu_number] <=
      flowprobe_get_headersize ())
    return;

  u32 i, index = vec_len (frm->streams);
  for (i = 0; i < index; i++)
    if (frm->streams[i].domain_id == 1)
      {
	index = i;
	break;
      }
  if (i == vec_len (frm->streams))
    {
      vec_validate (frm->streams, index);
      frm->streams[index].domain_id = 1;
    }
  stream = &frm->streams[index];

  tp = vlib_buffer_get_current (b0);
  ip = (ip4_header_t *) & tp->ip4;
  udp = (udp_header_t *) (ip + 1);
  h = (ipfix_message_header_t *) (udp + 1);
  s = (ipfix_set_header_t *) (h + 1);

  ip->ip_version_and_header_length = 0x45;
  ip->ttl = 254;
  ip->protocol = IP_PROTOCOL_UDP;
  ip->flags_and_fragment_offset = 0;
  ip->src_address.as_u32 = frm->src_address.as_u32;
  ip->dst_address.as_u32 = frm->ipfix_collector.as_u32;
  udp->src_port = clib_host_to_net_u16 (UDP_DST_PORT_ipfix);
  udp->dst_port = clib_host_to_net_u16 (UDP_DST_PORT_ipfix);
  udp->checksum = 0;

  /* FIXUP: message header export_time */
  h->export_time = (u32)
    (((f64) frm->unix_time_0) +
     (vlib_time_now (frm->vlib_main) - frm->vlib_time_0));
  h->export_time = clib_host_to_net_u32 (h->export_time);
  h->domain_id = clib_host_to_net_u32 (stream->domain_id);

  /* FIXUP: message header sequence_number */
  h->sequence_number = stream->sequence_number++;
  h->sequence_number = clib_host_to_net_u32 (h->sequence_number);

  s->set_id_length = ipfix_set_id_length (fm->template_reports[flags],
					  b0->current_length -
					  (sizeof (*ip) + sizeof (*udp) +
					   sizeof (*h)));
  h->version_length = version_length (b0->current_length -
				      (sizeof (*ip) + sizeof (*udp)));

  ip->length = clib_host_to_net_u16 (b0->current_length);

  ip->checksum = ip4_header_checksum (ip);
  udp->length = clib_host_to_net_u16 (b0->current_length - sizeof (*ip));

  if (frm->udp_checksum)
    {
      /* RFC 7011 section 10.3.2. */
      udp->checksum = ip4_tcp_udp_compute_checksum (vm, b0, ip);
      if (udp->checksum == 0)
	udp->checksum = 0xffff;
    }

  ASSERT (ip->checksum == ip4_header_checksum (ip));

  /* Find or allocate a frame */
  f = fm->context[which].frames_per_worker[my_cpu_number];
  if (PREDICT_FALSE (f == 0))
    {
      u32 *to_next;
      f = vlib_get_frame_to_node (vm, ip4_lookup_node.index);
      fm->context[which].frames_per_worker[my_cpu_number] = f;
      u32 bi0 = vlib_get_buffer_index (vm, b0);

      /* Enqueue the buffer */
      to_next = vlib_frame_vector_args (f);
      to_next[0] = bi0;
      f->n_vectors = 1;
    }

  vlib_put_frame_to_node (vm, ip4_lookup_node.index, f);
  vlib_node_increment_counter (vm, flowprobe_l2_node.index,
			       FLOWPROBE_ERROR_EXPORTED_PACKETS, 1);

  fm->context[which].frames_per_worker[my_cpu_number] = 0;
  fm->context[which].buffers_per_worker[my_cpu_number] = 0;
  fm->context[which].next_record_offset_per_worker[my_cpu_number] =
    flowprobe_get_headersize ();
}

static vlib_buffer_t *
flowprobe_get_buffer (vlib_main_t * vm, flowprobe_variant_t which)
{
  flowprobe_main_t *fm = &flowprobe_main;
  flow_report_main_t *frm = &flow_report_main;
  vlib_buffer_t *b0;
  u32 bi0;
  vlib_buffer_free_list_t *fl;
  u32 my_cpu_number = vm->thread_index;

  /* Find or allocate a buffer */
  b0 = fm->context[which].buffers_per_worker[my_cpu_number];

  /* Need to allocate a buffer? */
  if (PREDICT_FALSE (b0 == 0))
    {
      if (vlib_buffer_alloc (vm, &bi0, 1) != 1)
	{
	  vlib_node_increment_counter (vm, flowprobe_l2_node.index,
				       FLOWPROBE_ERROR_BUFFER, 1);
	  return 0;
	}

      /* Initialize the buffer */
      b0 = fm->context[which].buffers_per_worker[my_cpu_number] =
	vlib_get_buffer (vm, bi0);
      fl =
	vlib_buffer_get_free_list (vm, VLIB_BUFFER_DEFAULT_FREE_LIST_INDEX);
      vlib_buffer_init_for_free_list (b0, fl);
      VLIB_BUFFER_TRACE_TRAJECTORY_INIT (b0);

      b0->current_data = 0;
      b0->current_length = flowprobe_get_headersize ();
      b0->flags |= (VLIB_BUFFER_TOTAL_LENGTH_VALID | VLIB_BUFFER_FLOW_REPORT);
      vnet_buffer (b0)->sw_if_index[VLIB_RX] = 0;
      vnet_buffer (b0)->sw_if_index[VLIB_TX] = frm->fib_index;
      fm->context[which].next_record_offset_per_worker[my_cpu_number] =
	b0->current_length;
    }

  return b0;
}

static void
flowprobe_export_entry (vlib_main_t * vm, flowprobe_entry_t * e)
{
  u32 my_cpu_number = vm->thread_index;
  flowprobe_main_t *fm = &flowprobe_main;
  flow_report_main_t *frm = &flow_report_main;
  vlib_buffer_t *b0;
  bool collect_ip4 = false, collect_ip6 = false;
  flowprobe_variant_t which = e->key.which;
  flowprobe_record_t flags = fm->context[which].flags;
  u16 offset =
    fm->context[which].next_record_offset_per_worker[my_cpu_number];

  if (offset < flowprobe_get_headersize ())
    offset = flowprobe_get_headersize ();

  b0 = flowprobe_get_buffer (vm, which);
  /* No available buffer, what to do... */
  if (b0 == 0)
    return;

  if (flags & FLOW_RECORD_L3)
    {
      collect_ip4 = which == FLOW_VARIANT_L2_IP4 || which == FLOW_VARIANT_IP4;
      collect_ip6 = which == FLOW_VARIANT_L2_IP6 || which == FLOW_VARIANT_IP6;
    }

  offset += flowprobe_common_add (b0, e, offset);

  if (flags & FLOW_RECORD_L2)
    offset += flowprobe_l2_add (b0, e, offset);
  if (collect_ip6)
    offset += flowprobe_l3_ip6_add (b0, e, offset);
  if (collect_ip4)
    offset += flowprobe_l3_ip4_add (b0, e, offset);
  if (flags & FLOW_RECORD_L4)
    offset += flowprobe_l4_add (b0, e, offset);

  /* Reset per flow-export counters */
  e->packetcount = 0;
  e->octetcount = 0;
  e->last_exported = vlib_time_now (vm);

  b0->current_length = offset;

  fm->context[which].next_record_offset_per_worker[my_cpu_number] = offset;
  /* Time to flush the buffer? */
  if (offset + fm->template_size[flags] > frm->path_mtu)
    flowprobe_export_send (vm, b0, which);
}

uword
flowprobe_node_fn (vlib_main_t * vm,
		   vlib_node_runtime_t * node, vlib_frame_t * frame,
		   flowprobe_variant_t which)
{
  u32 n_left_from, *from, *to_next;
  flowprobe_next_t next_index;
  flowprobe_main_t *fm = &flowprobe_main;
  u64 now;

  now = (u64) ((vlib_time_now (vm) - fm->vlib_time_0) * 1e9);
  now += fm->nanosecond_time_0;

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

  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 >= 4 && n_left_to_next >= 2)
	{
	  u32 next0 = FLOWPROBE_NEXT_DROP;
	  u32 next1 = FLOWPROBE_NEXT_DROP;
	  u16 len0, len1;
	  u32 bi0, bi1;
	  vlib_buffer_t *b0, *b1;

	  /* Prefetch next iteration. */
	  {
	    vlib_buffer_t *p2, *p3;

	    p2 = vlib_get_buffer (vm, from[2]);
	    p3 = vlib_get_buffer (vm, from[3]);

	    vlib_prefetch_buffer_header (p2, LOAD);
	    vlib_prefetch_buffer_header (p3, LOAD);

	    CLIB_PREFETCH (p2->data, CLIB_CACHE_LINE_BYTES, STORE);
	    CLIB_PREFETCH (p3->data, CLIB_CACHE_LINE_BYTES, STORE);
	  }

	  /* speculatively enqueue b0 and b1 to the current next frame */
	  to_next[0] = bi0 = from[0];
	  to_next[1] = bi1 = from[1];
	  from += 2;
	  to_next += 2;
	  n_left_from -= 2;
	  n_left_to_next -= 2;

	  b0 = vlib_get_buffer (vm, bi0);
	  b1 = vlib_get_buffer (vm, bi1);

	  vnet_feature_next (vnet_buffer (b0)->sw_if_index[VLIB_TX],
			     &next0, b0);
	  vnet_feature_next (vnet_buffer (b1)->sw_if_index[VLIB_TX],
			     &next1, b1);

	  len0 = vlib_buffer_length_in_chain (vm, b0);
	  ethernet_header_t *eh0 = vlib_buffer_get_current (b0);
	  u16 ethertype0 = clib_net_to_host_u16 (eh0->type);

	  if (PREDICT_TRUE ((b0->flags & VLIB_BUFFER_FLOW_REPORT) == 0))
	    add_to_flow_record_state (vm, node, fm, b0, now, len0,
				      flowprobe_get_variant
				      (which, fm->context[which].flags,
				       ethertype0), 0);

	  len1 = vlib_buffer_length_in_chain (vm, b1);
	  ethernet_header_t *eh1 = vlib_buffer_get_current (b1);
	  u16 ethertype1 = clib_net_to_host_u16 (eh1->type);

	  if (PREDICT_TRUE ((b1->flags & VLIB_BUFFER_FLOW_REPORT) == 0))
	    add_to_flow_record_state (vm, node, fm, b1, now, len1,
				      flowprobe_get_variant
				      (which, fm->context[which].flags,
				       ethertype1), 0);

	  /* verify speculative enqueues, maybe switch current next frame */
	  vlib_validate_buffer_enqueue_x2 (vm, node, next_index,
					   to_next, n_left_to_next,
					   bi0, bi1, next0, next1);
	}

      while (n_left_from > 0 && n_left_to_next > 0)
	{
	  u32 bi0;
	  vlib_buffer_t *b0;
	  u32 next0 = FLOWPROBE_NEXT_DROP;
	  u16 len0;

	  /* speculatively enqueue b0 to the current next frame */
	  bi0 = from[0];
	  to_next[0] = bi0;
	  from += 1;
	  to_next += 1;
	  n_left_from -= 1;
	  n_left_to_next -= 1;

	  b0 = vlib_get_buffer (vm, bi0);

	  vnet_feature_next (vnet_buffer (b0)->sw_if_index[VLIB_TX],
			     &next0, b0);

	  len0 = vlib_buffer_length_in_chain (vm, b0);
	  ethernet_header_t *eh0 = vlib_buffer_get_current (b0);
	  u16 ethertype0 = clib_net_to_host_u16 (eh0->type);

	  if (PREDICT_TRUE ((b0->flags & VLIB_BUFFER_FLOW_REPORT) == 0))
	    {
	      flowprobe_trace_t *t = 0;
	      if (PREDICT_FALSE ((node->flags & VLIB_NODE_FLAG_TRACE)
				 && (b0->flags & VLIB_BUFFER_IS_TRACED)))
		t = vlib_add_trace (vm, node, b0, sizeof (*t));

	      add_to_flow_record_state (vm, node, fm, b0, now, len0,
					flowprobe_get_variant
					(which, fm->context[which].flags,
					 ethertype0), t);
	    }

	  /* verify speculative enqueue, maybe switch current next frame */
	  vlib_validate_buffer_enqueue_x1 (vm, node, next_index,
					   to_next, n_left_to_next,
					   bi0, next0);
	}

      vlib_put_next_frame (vm, node, next_index, n_left_to_next);
    }
  return frame->n_vectors;
}

static uword
flowprobe_ip4_node_fn (vlib_main_t * vm,
		       vlib_node_runtime_t * node, vlib_frame_t * frame)
{
  return flowprobe_node_fn (vm, node, frame, FLOW_VARIANT_IP4);
}

static uword
flowprobe_ip6_node_fn (vlib_main_t * vm,
		       vlib_node_runtime_t * node, vlib_frame_t * frame)
{
  return flowprobe_node_fn (vm, node, frame, FLOW_VARIANT_IP6);
}

static uword
flowprobe_l2_node_fn (vlib_main_t * vm,
		      vlib_node_runtime_t * node, vlib_frame_t * frame)
{
  return flowprobe_node_fn (vm, node, frame, FLOW_VARIANT_L2);
}

static inline void
flush_record (flowprobe_variant_t which)
{
  vlib_main_t *vm = vlib_get_main ();
  vlib_buffer_t *b = flowprobe_get_buffer (vm, which);
  if (b)
    flowprobe_export_send (vm, b, which);
}

void
flowprobe_flush_callback_ip4 (void)
{
  flush_record (FLOW_VARIANT_IP4);
}

void
flowprobe_flush_callback_ip6 (void)
{
  flush_record (FLOW_VARIANT_IP6);
}

void
flowprobe_flush_callback_l2 (void)
{
  flush_record (FLOW_VARIANT_L2);
  flush_record (FLOW_VARIANT_L2_IP4);
  flush_record (FLOW_VARIANT_L2_IP6);
}


static void
flowprobe_delete_by_index (u32 my_cpu_number, u32 poolindex)
{
  flowprobe_main_t *fm = &flowprobe_main;
  flowprobe_entry_t *e;
  u32 h;

  e = pool_elt_at_index (fm->pool_per_worker[my_cpu_number], poolindex);

  /* Get my index */
  h = flowprobe_hash (&e->key);

  /* Reset hash */
  fm->hash_per_worker[my_cpu_number][h] = ~0;

  pool_put_index (fm->pool_per_worker[my_cpu_number], poolindex);
}


/* Per worker process processing the active/passive expired entries */
static uword
flowprobe_walker_process (vlib_main_t * vm,
			  vlib_node_runtime_t * rt, vlib_frame_t * f)
{
  flowprobe_main_t *fm = &flowprobe_main;
  flow_report_main_t *frm = &flow_report_main;
  flowprobe_entry_t *e;

  /*
   * $$$$ Remove this check from here and track FRM status and disable
   * this process if required.
   */
  if (frm->ipfix_collector.as_u32 == 0 || frm->src_address.as_u32 == 0)
    {
      fm->disabled = true;
      return 0;
    }
  fm->disabled = false;

  u32 cpu_index = os_get_thread_index ();
  u32 *to_be_removed = 0, *i;

  /*
   * Tick the timer when required and process the vector of expired
   * timers
   */
  f64 start_time = vlib_time_now (vm);
  u32 count = 0;

  tw_timer_expire_timers_2t_1w_2048sl (fm->timers_per_worker[cpu_index],
				       start_time);

  vec_foreach (i, fm->expired_passive_per_worker[cpu_index])
  {
    u32 exported = 0;
    f64 now = vlib_time_now (vm);
    if (now > start_time + 100e-6
	|| exported > FLOW_MAXIMUM_EXPORT_ENTRIES - 1)
      break;

    if (pool_is_free_index (fm->pool_per_worker[cpu_index], *i))
      {
	clib_warning ("Element is %d is freed already\n", *i);
	continue;
      }
    else
      e = pool_elt_at_index (fm->pool_per_worker[cpu_index], *i);

    /* Check last update timestamp. If it is longer than passive time nuke
     * entry. Otherwise restart timer with what's left
     * Premature passive timer by more than 10%
     */
    if ((now - e->last_updated) < (fm->passive_timer * 0.9))
      {
	f64 delta = fm->passive_timer - (now - e->last_updated);
	e->passive_timer_handle = tw_timer_start_2t_1w_2048sl
	  (fm->timers_per_worker[cpu_index], *i, 0, delta);
      }
    else			/* Nuke entry */
      {
	vec_add1 (to_be_removed, *i);
      }
    /* If anything to report send it to the exporter */
    if (e->packetcount && now > e->last_exported + fm->active_timer)
      {
	exported++;
	flowprobe_export_entry (vm, e);
      }
    count++;
  }
  if (count)
    vec_delete (fm->expired_passive_per_worker[cpu_index], count, 0);

  vec_foreach (i, to_be_removed) flowprobe_delete_by_index (cpu_index, *i);
  vec_free (to_be_removed);

  return 0;
}

/* *INDENT-OFF* */
VLIB_REGISTER_NODE (flowprobe_ip4_node) = {
  .function = flowprobe_ip4_node_fn,
  .name = "flowprobe-ip4",
  .vector_size = sizeof (u32),
  .format_trace = format_flowprobe_trace,
  .type = VLIB_NODE_TYPE_INTERNAL,
  .n_errors = ARRAY_LEN(flowprobe_error_strings),
  .error_strings = flowprobe_error_strings,
  .n_next_nodes = FLOWPROBE_N_NEXT,
  .next_nodes = FLOWPROBE_NEXT_NODES,
};
VLIB_REGISTER_NODE (flowprobe_ip6_node) = {
  .function = flowprobe_ip6_node_fn,
  .name = "flowprobe-ip6",
  .vector_size = sizeof (u32),
  .format_trace = format_flowprobe_trace,
  .type = VLIB_NODE_TYPE_INTERNAL,
  .n_errors = ARRAY_LEN(flowprobe_error_strings),
  .error_strings = flowprobe_error_strings,
  .n_next_nodes = FLOWPROBE_N_NEXT,
  .next_nodes = FLOWPROBE_NEXT_NODES,
};
VLIB_REGISTER_NODE (flowprobe_l2_node) = {
  .function = flowprobe_l2_node_fn,
  .name = "flowprobe-l2",
  .vector_size = sizeof (u32),
  .format_trace = format_flowprobe_trace,
  .type = VLIB_NODE_TYPE_INTERNAL,
  .n_errors = ARRAY_LEN(flowprobe_error_strings),
  .error_strings = flowprobe_error_strings,
  .n_next_nodes = FLOWPROBE_N_NEXT,
  .next_nodes = FLOWPROBE_NEXT_NODES,
};
VLIB_REGISTER_NODE (flowprobe_walker_node) = {
  .function = flowprobe_walker_process,
  .name = "flowprobe-walker",
  .type = VLIB_NODE_TYPE_INPUT,
  .state = VLIB_NODE_STATE_INTERRUPT,
};
/* *INDENT-ON* */

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