summaryrefslogtreecommitdiffstats
path: root/src/vat2
diff options
context:
space:
mode:
authorMiklos Tirpak <miklos.tirpak@gmail.com>2021-01-12 15:14:02 +0100
committerOle Tr�an <otroan@employees.org>2021-02-04 09:56:54 +0000
commit38fae310843b7431136f40bfa8cf7c6bec59450f (patch)
tree6a45176f8a5782f23bc4a579a14d3ad3646e7d37 /src/vat2
parent15cb3dac874bcec701a33723467b81a53a6fc4d3 (diff)
vlib: increase the stats epoch only when necessary
When the counter vectors are validated and they are already long enough to fit the given index in memory, there is no need to increase the stats segment epoch. In this case, the counter vectors do not change as a result of the validation. This optimization is necessary for the case when the configuration is changed at multiple thousands per second rate. The counter vectors grow at the beginning and their size stabilizes after a while. Without this improvement, it can still take several seconds for a stats reader to succeed. Type: improvement Signed-off-by: Miklos Tirpak <miklos.tirpak@gmail.com> Change-Id: I5a6c30255832716a1460018d0bd0f63031de102b
Diffstat (limited to 'src/vat2')
0 files changed, 0 insertions, 0 deletions
0' href='#n140'>140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 
/*
 * ipsec_output.c : IPSec output node
 *
 * Copyright (c) 2015 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/vnet.h>
#include <vnet/api_errno.h>
#include <vnet/ip/ip.h>

#include <vnet/ipsec/ipsec.h>
#include <vnet/ipsec/ipsec_io.h>

#if WITH_LIBSSL > 0

#define foreach_ipsec_output_error                   \
 _(RX_PKTS, "IPSec pkts received")                   \
 _(POLICY_DISCARD, "IPSec policy discard")           \
 _(POLICY_NO_MATCH, "IPSec policy (no match)")       \
 _(POLICY_PROTECT, "IPSec policy protect")           \
 _(POLICY_BYPASS, "IPSec policy bypass")             \
 _(ENCAPS_FAILED, "IPSec encapsulation failed")

typedef enum
{
#define _(sym,str) IPSEC_OUTPUT_ERROR_##sym,
  foreach_ipsec_output_error
#undef _
    IPSEC_DECAP_N_ERROR,
} ipsec_output_error_t;

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

typedef struct
{
  u32 spd_id;
  u32 policy_id;
} ipsec_output_trace_t;

/* packet trace format function */
static u8 *
format_ipsec_output_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 *);
  ipsec_output_trace_t *t = va_arg (*args, ipsec_output_trace_t *);

  s = format (s, "spd %u policy %d", t->spd_id, t->policy_id);

  return s;
}

always_inline ipsec_policy_t *
ipsec_output_policy_match (ipsec_spd_t * spd, u8 pr, u32 la, u32 ra, u16 lp,
			   u16 rp)
{
  ipsec_main_t *im = &ipsec_main;
  ipsec_policy_t *p;
  u32 *i;

  if (!spd)
    return 0;

  vec_foreach (i, spd->policies[IPSEC_SPD_POLICY_IP4_OUTBOUND])
  {
    p = pool_elt_at_index (im->policies, *i);
    if (PREDICT_FALSE (p->protocol && (p->protocol != pr)))
      continue;

    if (ra < clib_net_to_host_u32 (p->raddr.start.ip4.as_u32))
      continue;

    if (ra > clib_net_to_host_u32 (p->raddr.stop.ip4.as_u32))
      continue;

    if (la < clib_net_to_host_u32 (p->laddr.start.ip4.as_u32))
      continue;

    if (la > clib_net_to_host_u32 (p->laddr.stop.ip4.as_u32))
      continue;

    if (PREDICT_FALSE
	((pr != IP_PROTOCOL_TCP) && (pr != IP_PROTOCOL_UDP)
	 && (pr != IP_PROTOCOL_SCTP)))
      return p;

    if (lp < p->lport.start)
      continue;

    if (lp > p->lport.stop)
      continue;

    if (rp < p->rport.start)
      continue;

    if (rp > p->rport.stop)
      continue;

    return p;
  }
  return 0;
}

always_inline uword
ip6_addr_match_range (ip6_address_t * a, ip6_address_t * la,
		      ip6_address_t * ua)
{
  if ((memcmp (a->as_u64, la->as_u64, 2 * sizeof (u64)) >= 0) &&
      (memcmp (a->as_u64, ua->as_u64, 2 * sizeof (u64)) <= 0))
    return 1;
  return 0;
}

always_inline ipsec_policy_t *
ipsec6_output_policy_match (ipsec_spd_t * spd,
			    ip6_address_t * la,
			    ip6_address_t * ra, u16 lp, u16 rp, u8 pr)
{
  ipsec_main_t *im = &ipsec_main;
  ipsec_policy_t *p;
  u32 *i;

  if (!spd)
    return 0;

  vec_foreach (i, spd->policies[IPSEC_SPD_POLICY_IP6_OUTBOUND])
  {
    p = pool_elt_at_index (im->policies, *i);
    if (PREDICT_FALSE (p->protocol && (p->protocol != pr)))
      continue;

    if (!ip6_addr_match_range (ra, &p->raddr.start.ip6, &p->raddr.stop.ip6))
      continue;

    if (!ip6_addr_match_range (la, &p->laddr.start.ip6, &p->laddr.stop.ip6))
      continue;

    if (PREDICT_FALSE
	((pr != IP_PROTOCOL_TCP) && (pr != IP_PROTOCOL_UDP)
	 && (pr != IP_PROTOCOL_SCTP)))
      return p;

    if (lp < p->lport.start)
      continue;

    if (lp > p->lport.stop)
      continue;

    if (rp < p->rport.start)
      continue;

    if (rp > p->rport.stop)
      continue;

    return p;
  }

  return 0;
}

static inline uword
ipsec_output_inline (vlib_main_t * vm, vlib_node_runtime_t * node,
		     vlib_frame_t * from_frame, int is_ipv6)
{
  ipsec_main_t *im = &ipsec_main;

  u32 *from, *to_next = 0, thread_index;
  u32 n_left_from, sw_if_index0, last_sw_if_index = (u32) ~ 0;
  u32 next_node_index = (u32) ~ 0, last_next_node_index = (u32) ~ 0;
  vlib_frame_t *f = 0;
  u32 spd_index0 = ~0;
  ipsec_spd_t *spd0 = 0;
  int bogus;
  u64 nc_protect = 0, nc_bypass = 0, nc_discard = 0, nc_nomatch = 0;

  from = vlib_frame_vector_args (from_frame);
  n_left_from = from_frame->n_vectors;
  thread_index = vm->thread_index;

  while (n_left_from > 0)
    {
      u32 bi0, pi0, bi1;
      vlib_buffer_t *b0, *b1;
      ipsec_policy_t *p0;
      ip4_header_t *ip0;
      ip6_header_t *ip6_0 = 0;
      udp_header_t *udp0;
      u32 iph_offset = 0;
      tcp_header_t *tcp0;
      u64 bytes0;

      bi0 = from[0];
      b0 = vlib_get_buffer (vm, bi0);
      if (n_left_from > 1)
	{
	  bi1 = from[1];
	  b1 = vlib_get_buffer (vm, bi1);
	  CLIB_PREFETCH (b1, CLIB_CACHE_LINE_BYTES * 2, STORE);
	  vlib_prefetch_buffer_data (b1, LOAD);
	}
      sw_if_index0 = vnet_buffer (b0)->sw_if_index[VLIB_TX];
      iph_offset = vnet_buffer (b0)->ip.save_rewrite_length;
      ip0 = (ip4_header_t *) ((u8 *) vlib_buffer_get_current (b0)
			      + iph_offset);

      /* lookup for SPD only if sw_if_index is changed */
      if (PREDICT_FALSE (last_sw_if_index != sw_if_index0))
	{
	  uword *p = hash_get (im->spd_index_by_sw_if_index, sw_if_index0);
	  ASSERT (p);
	  spd_index0 = p[0];
	  spd0 = pool_elt_at_index (im->spds, spd_index0);
	  last_sw_if_index = sw_if_index0;
	}

      if (is_ipv6)
	{
	  ip6_0 = (ip6_header_t *) ((u8 *) vlib_buffer_get_current (b0)
				    + iph_offset);

	  udp0 = ip6_next_header (ip6_0);
#if 0
	  clib_warning
	    ("packet received from %U port %u to %U port %u spd_id %u",
	     format_ip6_address, &ip6_0->src_address,
	     clib_net_to_host_u16 (udp0->src_port), format_ip6_address,
	     &ip6_0->dst_address, clib_net_to_host_u16 (udp0->dst_port),
	     spd0->id);
#endif

	  p0 = ipsec6_output_policy_match (spd0,
					   &ip6_0->src_address,
					   &ip6_0->dst_address,
					   clib_net_to_host_u16
					   (udp0->src_port),
					   clib_net_to_host_u16
					   (udp0->dst_port), ip6_0->protocol);
	}
      else
	{
	  udp0 = (udp_header_t *) ((u8 *) ip0 + ip4_header_bytes (ip0));

#if 0
	  clib_warning ("packet received from %U to %U port %u",
			format_ip4_address, ip0->src_address.as_u8,
			format_ip4_address, ip0->dst_address.as_u8,
			clib_net_to_host_u16 (udp0->dst_port));
	  clib_warning ("sw_if_index0 %u spd_index0 %u spd_id %u",
			sw_if_index0, spd_index0, spd0->id);
#endif

	  p0 = ipsec_output_policy_match (spd0, ip0->protocol,
					  clib_net_to_host_u32
					  (ip0->src_address.as_u32),
					  clib_net_to_host_u32
					  (ip0->dst_address.as_u32),
					  clib_net_to_host_u16
					  (udp0->src_port),
					  clib_net_to_host_u16
					  (udp0->dst_port));
	}
      tcp0 = (void *) udp0;

      if (PREDICT_TRUE (p0 != NULL))
	{
	  pi0 = p0 - im->policies;

	  vlib_prefetch_combined_counter (&ipsec_spd_policy_counters,
					  thread_index, pi0);

	  if (is_ipv6)
	    {
	      bytes0 = clib_net_to_host_u16 (ip6_0->payload_length);
	      bytes0 += sizeof (ip6_header_t);
	    }
	  else
	    {
	      bytes0 = clib_net_to_host_u16 (ip0->length);
	    }

	  if (p0->policy == IPSEC_POLICY_ACTION_PROTECT)
	    {
	      ipsec_sa_t *sa = 0;
	      nc_protect++;
	      sa = pool_elt_at_index (im->sad, p0->sa_index);
	      if (sa->protocol == IPSEC_PROTOCOL_ESP)
		if (is_ipv6)
		  next_node_index = im->esp6_encrypt_node_index;
		else
		  next_node_index = im->esp4_encrypt_node_index;
	      else if (is_ipv6)
		next_node_index = im->ah6_encrypt_node_index;
	      else
		next_node_index = im->ah4_encrypt_node_index;
	      vnet_buffer (b0)->ipsec.sad_index = p0->sa_index;

	      if (is_ipv6)
		{
		  if (PREDICT_FALSE
		      (b0->flags & VNET_BUFFER_F_OFFLOAD_TCP_CKSUM))
		    {
		      tcp0->checksum =
			ip6_tcp_udp_icmp_compute_checksum (vm, b0, ip6_0,
							   &bogus);
		      b0->flags &= ~VNET_BUFFER_F_OFFLOAD_TCP_CKSUM;
		    }
		  if (PREDICT_FALSE
		      (b0->flags & VNET_BUFFER_F_OFFLOAD_UDP_CKSUM))
		    {
		      udp0->checksum =
			ip6_tcp_udp_icmp_compute_checksum (vm, b0, ip6_0,
							   &bogus);
		      b0->flags &= ~VNET_BUFFER_F_OFFLOAD_UDP_CKSUM;
		    }
		}
	      else
		{
		  if (b0->flags & VNET_BUFFER_F_OFFLOAD_IP_CKSUM)
		    {
		      ip0->checksum = ip4_header_checksum (ip0);
		      b0->flags &= ~VNET_BUFFER_F_OFFLOAD_IP_CKSUM;
		    }
		  if (PREDICT_FALSE
		      (b0->flags & VNET_BUFFER_F_OFFLOAD_TCP_CKSUM))
		    {
		      tcp0->checksum =
			ip4_tcp_udp_compute_checksum (vm, b0, ip0);
		      b0->flags &= ~VNET_BUFFER_F_OFFLOAD_TCP_CKSUM;
		    }
		  if (PREDICT_FALSE
		      (b0->flags & VNET_BUFFER_F_OFFLOAD_UDP_CKSUM))
		    {
		      udp0->checksum =
			ip4_tcp_udp_compute_checksum (vm, b0, ip0);
		      b0->flags &= ~VNET_BUFFER_F_OFFLOAD_UDP_CKSUM;
		    }
		}
	      vlib_buffer_advance (b0, iph_offset);
	    }
	  else if (p0->policy == IPSEC_POLICY_ACTION_BYPASS)
	    {
	      nc_bypass++;
	      next_node_index = get_next_output_feature_node_index (b0, node);
	    }
	  else
	    {
	      nc_discard++;
	      next_node_index = im->error_drop_node_index;
	    }
	  vlib_increment_combined_counter
	    (&ipsec_spd_policy_counters, thread_index, pi0, 1, bytes0);
	}
      else
	{
	  pi0 = ~0;
	  nc_nomatch++;
	  next_node_index = im->error_drop_node_index;
	}

      from += 1;
      n_left_from -= 1;

      if (PREDICT_FALSE ((last_next_node_index != next_node_index) || f == 0))
	{
	  /* if this is not 1st frame */
	  if (f)
	    vlib_put_frame_to_node (vm, last_next_node_index, f);

	  last_next_node_index = next_node_index;

	  f = vlib_get_frame_to_node (vm, next_node_index);

	  /* frame->frame_flags, copy it from node */
	  /* Copy trace flag from next_frame and from runtime. */
	  f->frame_flags |= node->flags & VLIB_NODE_FLAG_TRACE;

	  to_next = vlib_frame_vector_args (f);
	}

      to_next[0] = bi0;
      to_next += 1;
      f->n_vectors++;

      if (PREDICT_FALSE (node->flags & VLIB_NODE_FLAG_TRACE) &&
	  PREDICT_FALSE (b0->flags & VLIB_BUFFER_IS_TRACED))
	{
	  ipsec_output_trace_t *tr =
	    vlib_add_trace (vm, node, b0, sizeof (*tr));
	  if (spd0)
	    tr->spd_id = spd0->id;
	  tr->policy_id = pi0;
	}
    }

  vlib_put_frame_to_node (vm, next_node_index, f);
  vlib_node_increment_counter (vm, node->node_index,
			       IPSEC_OUTPUT_ERROR_POLICY_PROTECT, nc_protect);
  vlib_node_increment_counter (vm, node->node_index,
			       IPSEC_OUTPUT_ERROR_POLICY_BYPASS, nc_bypass);
  vlib_node_increment_counter (vm, node->node_index,
			       IPSEC_OUTPUT_ERROR_POLICY_DISCARD, nc_discard);
  vlib_node_increment_counter (vm, node->node_index,
			       IPSEC_OUTPUT_ERROR_POLICY_NO_MATCH,
			       nc_nomatch);
  return from_frame->n_vectors;
}

VLIB_NODE_FN (ipsec4_output_node) (vlib_main_t * vm,
				   vlib_node_runtime_t * node,
				   vlib_frame_t * frame)
{
  return ipsec_output_inline (vm, node, frame, 0);
}

/* *INDENT-OFF* */
VLIB_REGISTER_NODE (ipsec4_output_node) = {
  .name = "ipsec4-output-feature",
  .vector_size = sizeof (u32),
  .format_trace = format_ipsec_output_trace,
  .type = VLIB_NODE_TYPE_INTERNAL,

  .n_errors = ARRAY_LEN(ipsec_output_error_strings),
  .error_strings = ipsec_output_error_strings,

  .n_next_nodes = IPSEC_OUTPUT_N_NEXT,
  .next_nodes = {
#define _(s,n) [IPSEC_OUTPUT_NEXT_##s] = n,
    foreach_ipsec_output_next
#undef _
  },
};
/* *INDENT-ON* */

VLIB_NODE_FN (ipsec6_output_node) (vlib_main_t * vm,
				   vlib_node_runtime_t * node,
				   vlib_frame_t * frame)
{
  return ipsec_output_inline (vm, node, frame, 1);
}

/* *INDENT-OFF* */
VLIB_REGISTER_NODE (ipsec6_output_node) = {
  .name = "ipsec6-output-feature",
  .vector_size = sizeof (u32),
  .format_trace = format_ipsec_output_trace,
  .type = VLIB_NODE_TYPE_INTERNAL,

  .n_errors = ARRAY_LEN(ipsec_output_error_strings),
  .error_strings = ipsec_output_error_strings,

  .n_next_nodes = IPSEC_OUTPUT_N_NEXT,
  .next_nodes = {
#define _(s,n) [IPSEC_OUTPUT_NEXT_##s] = n,
    foreach_ipsec_output_next
#undef _
  },
};
/* *INDENT-ON* */

#else /* IPSEC > 1 */

/* Dummy ipsec output node, in case when IPSec is disabled */

static uword
ipsec_output_node_fn (vlib_main_t * vm,
		      vlib_node_runtime_t * node, vlib_frame_t * frame)
{
  return 0;
}

/* *INDENT-OFF* */
VLIB_REGISTER_NODE (ipsec4_output_node) = {
  .vector_size = sizeof (u32),
  .function = ipsec_output_node_fn,
  .name = "ipsec4-output-feature",
};

VLIB_REGISTER_NODE (ipsec6_output_node) = {
  .vector_size = sizeof (u32),
  .function = ipsec_output_node_fn,
  .name = "ipsec6-output-feature",
};
/* *INDENT-ON* */
#endif

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