summaryrefslogtreecommitdiffstats
path: root/src/plugins/igmp/igmp_query.c
diff options
context:
space:
mode:
authorMarvin Liu <yong.liu@intel.com>2022-08-17 09:38:40 +0800
committerDamjan Marion <dmarion@0xa5.net>2022-08-25 19:05:40 +0000
commitabd5669422c5805da5135496d5e5a394fa5aa602 (patch)
treea464eb14b5e04b19042e92bb83ca7b8567731f19 /src/plugins/igmp/igmp_query.c
parent9a6ad01c0d443f002eafa709813d021bf0c98eac (diff)
vlib: introduce DMA infrastructure
This patch introduces DMA infrastructure into vlib. This is well known that large amount of memory movements will drain core resource. Nowadays more and more hardware accelerators were designed out for freeing core from this burden. Meanwhile some restrictions still remained when utilizing hardware accelerators, e.g. cross numa throughput will have a significant drop compared to same node. Normally the number of hardware accelerator instances will less than cores number, not to mention that applications number will even beyond the number of cores. Some hardware may support share virtual address with cores, while others are not. Here we introduce new DMA infrastructure which can fulfill the requirements of vpp applications like session and memif and in the meantime dealing with hardware limitations. Here is some design backgrounds: Backend is the abstract of resource which allocated from DMA device and can do some basic operations like configuration, DMA copy and result query. Config is the abstract of application DMA requirement. Application need to request an unique config index from DMA infrastructure. This unique config index is associated with backend resource. Two options cpu fallback and barrier before last can be specified in config. DMA transfer will be performed by CPU when backend is busy if cpu fallback option is enabled. DMA transfer callback will be in order if barrier before last option is enabled. We constructs all the stuffs that DMA transfer request needed into DMA batch. It contains the pattern of DMA descriptors and function pointers for submission and callback. One DMA transfer request need multiple times batch update and one time batch submission. DMA backends will assigned to config's workers threads equally. Lock will be used for thread-safety if same backends assigned to multiple threads. Backend node will check all the pending requests in worker thread and do callback with the pointer of DMA batch if transfer completed. Application can utilize cookie in DMA batch for selves usage. DMA architecture: +----------+ +----------+ +----------+ +----------+ | Config1 | | Config2 | | Config1 | | Config2 | +----------+ +----------+ +----------+ +----------+ || || || || +-------------------------+ +-------------------------+ | DMA polling thread A | | DMA polling thread B | +-------------------------+ +-------------------------+ || || +----------+ +----------+ | Backend1 | | Backend2 | +----------+ +----------+ Type: feature Signed-off-by: Marvin Liu <yong.liu@intel.com> Change-Id: I1725e0c26687985aac29618c9abe4f5e0de08ebf
Diffstat (limited to 'src/plugins/igmp/igmp_query.c')
0 files changed, 0 insertions, 0 deletions
id='n228' href='#n228'>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 
/*
 * decap.c : L2TPv3 tunnel decapsulation
 *
 * Copyright (c) 2013 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 <vppinfra/error.h>
#include <vppinfra/hash.h>
#include <vnet/vnet.h>
#include <vnet/ip/ip.h>
#include <vnet/ethernet/ethernet.h>
#include <vnet/l2tp/l2tp.h>
#include <vnet/l2/l2_input.h>

/* Statistics (not really errors) */
#define foreach_l2t_decap_error                                 \
_(USER_TO_NETWORK, "L2TP user (ip6) to L2 network pkts")        \
_(SESSION_ID_MISMATCH, "l2tpv3 local session id mismatches")    \
_(COOKIE_MISMATCH, "l2tpv3 local cookie mismatches")            \
_(NO_SESSION, "l2tpv3 session not found")                       \
_(ADMIN_DOWN, "l2tpv3 tunnel is down")

static char *l2t_decap_error_strings[] = {
#define _(sym,string) string,
  foreach_l2t_decap_error
#undef _
};

typedef enum
{
#define _(sym,str) L2T_DECAP_ERROR_##sym,
  foreach_l2t_decap_error
#undef _
    L2T_DECAP_N_ERROR,
} l2t_DECAP_error_t;

typedef enum
{
  L2T_DECAP_NEXT_DROP,
  L2T_DECAP_NEXT_L2_INPUT,
  L2T_DECAP_N_NEXT,
  /* Pseudo next index */
  L2T_DECAP_NEXT_NO_INTERCEPT = L2T_DECAP_N_NEXT,
} l2t_decap_next_t;

#define NSTAGES 3

static inline void
stage0 (vlib_main_t * vm, vlib_node_runtime_t * node, vlib_buffer_t * b)
{
  vlib_prefetch_buffer_header (b, STORE);
  /* l2tpv3 header is a long way away, need 2 cache lines */
  CLIB_PREFETCH (b->data, 2 * CLIB_CACHE_LINE_BYTES, STORE);
}

static inline void
stage1 (vlib_main_t * vm, vlib_node_runtime_t * node, vlib_buffer_t * b)
{
  l2t_main_t *lm = &l2t_main;
  ip6_header_t *ip6 = vlib_buffer_get_current (b);
  u32 session_index;
  uword *p = 0;
  l2tpv3_header_t *l2t;

  /* Not L2tpv3 (0x73, 0t115)? Use the normal path. */
  if (PREDICT_FALSE (ip6->protocol != IP_PROTOCOL_L2TP))
    {
      vnet_buffer (b)->l2t.next_index = L2T_DECAP_NEXT_NO_INTERCEPT;
      return;
    }

  /* Make up your minds, people... */
  switch (lm->lookup_type)
    {
    case L2T_LOOKUP_SRC_ADDRESS:
      p = hash_get_mem (lm->session_by_src_address, &ip6->src_address);
      break;
    case L2T_LOOKUP_DST_ADDRESS:
      p = hash_get_mem (lm->session_by_dst_address, &ip6->dst_address);
      break;
    case L2T_LOOKUP_SESSION_ID:
      l2t = (l2tpv3_header_t *) (ip6 + 1);
      p = hash_get (lm->session_by_session_id, l2t->session_id);
      break;
    default:
      ASSERT (0);
    }

  if (PREDICT_FALSE (p == 0))
    {
      vnet_buffer (b)->l2t.next_index = L2T_DECAP_NEXT_NO_INTERCEPT;
      return;
    }
  else
    {
      session_index = p[0];
    }

  /* Remember mapping index, prefetch the mini counter */
  vnet_buffer (b)->l2t.next_index = L2T_DECAP_NEXT_L2_INPUT;
  vnet_buffer (b)->l2t.session_index = session_index;

  /* $$$$$ prefetch counter */
}

static inline u32
last_stage (vlib_main_t * vm, vlib_node_runtime_t * node, vlib_buffer_t * b)
{
  l2t_main_t *lm = &l2t_main;
  ip6_header_t *ip6 = vlib_buffer_get_current (b);
  vlib_node_t *n = vlib_get_node (vm, node->node_index);
  u32 node_counter_base_index = n->error_heap_index;
  vlib_error_main_t *em = &vm->error_main;
  l2tpv3_header_t *l2tp;
  u32 counter_index;
  l2t_session_t *session = 0;
  u32 session_index;
  u32 next_index;
  u8 l2tp_decap_local = (l2t_decap_local_node.index == n->index);

  /* Other-than-output pkt? We're done... */
  if (vnet_buffer (b)->l2t.next_index != L2T_DECAP_NEXT_L2_INPUT)
    {
      next_index = vnet_buffer (b)->l2t.next_index;
      goto done;
    }

  em->counters[node_counter_base_index + L2T_DECAP_ERROR_USER_TO_NETWORK] +=
    1;

  session_index = vnet_buffer (b)->l2t.session_index;

  counter_index =
    session_index_to_counter_index (session_index,
				    SESSION_COUNTER_USER_TO_NETWORK);

  /* per-mapping byte stats include the ethernet header */
  vlib_increment_combined_counter (&lm->counter_main,
				   vlib_get_thread_index (),
				   counter_index, 1 /* packet_increment */ ,
				   vlib_buffer_length_in_chain (vm, b) +
				   sizeof (ethernet_header_t));

  session = pool_elt_at_index (lm->sessions, session_index);

  l2tp = vlib_buffer_get_current (b) + sizeof (*ip6);

  if (PREDICT_FALSE (l2tp->session_id != session->local_session_id))
    {
      /* Key matched but session id does not. Assume packet is not for us. */
      em->counters[node_counter_base_index +
		   L2T_DECAP_ERROR_SESSION_ID_MISMATCH] += 1;
      next_index = L2T_DECAP_NEXT_NO_INTERCEPT;
      goto done;
    }

  if (PREDICT_FALSE (l2tp->cookie != session->local_cookie[0]))
    {
      if (l2tp->cookie != session->local_cookie[1])
	{
	  /* Key and session ID matched, but cookie doesn't. Drop this packet. */
	  b->error = node->errors[L2T_DECAP_ERROR_COOKIE_MISMATCH];
	  next_index = L2T_DECAP_NEXT_DROP;
	  goto done;
	}
    }

  vnet_buffer (b)->sw_if_index[VLIB_RX] = session->sw_if_index;

  if (PREDICT_FALSE (!(session->admin_up)))
    {
      b->error = node->errors[L2T_DECAP_ERROR_ADMIN_DOWN];
      next_index = L2T_DECAP_NEXT_DROP;
      goto done;
    }

  /* strip the ip6 and L2TP header */
  vlib_buffer_advance (b, sizeof (*ip6) + session->l2tp_hdr_size);

  /* Required to make the l2 tag push / pop code work on l2 subifs */
  vnet_update_l2_len (b);

  if (PREDICT_FALSE (b->flags & VLIB_BUFFER_IS_TRACED))
    {
      l2t_trace_t *t = vlib_add_trace (vm, node, b, sizeof (*t));
      t->is_user_to_network = 1;
      t->our_address.as_u64[0] = ip6->dst_address.as_u64[0];
      t->our_address.as_u64[1] = ip6->dst_address.as_u64[1];
      t->client_address.as_u64[0] = ip6->src_address.as_u64[0];
      t->client_address.as_u64[1] = ip6->src_address.as_u64[1];
      t->session_index = session_index;
    }

  return L2T_DECAP_NEXT_L2_INPUT;

done:
  if (next_index == L2T_DECAP_NEXT_NO_INTERCEPT)
    {
      /* Small behavioral change between l2tp-decap and l2tp-decap-local */
      if (l2tp_decap_local)
	{
	  b->error = node->errors[L2T_DECAP_ERROR_NO_SESSION];
	  next_index = L2T_DECAP_NEXT_DROP;
	}
      else
	{
	  /* Go to next node on the ip6 configuration chain */
	  if (PREDICT_TRUE (session != 0))
	    vnet_feature_next (&next_index, b);
	}
    }

  if (PREDICT_FALSE (b->flags & VLIB_BUFFER_IS_TRACED))
    {
      l2t_trace_t *t = vlib_add_trace (vm, node, b, sizeof (*t));
      t->is_user_to_network = 1;
      t->our_address.as_u64[0] = ip6->dst_address.as_u64[0];
      t->our_address.as_u64[1] = ip6->dst_address.as_u64[1];
      t->client_address.as_u64[0] = ip6->src_address.as_u64[0];
      t->client_address.as_u64[1] = ip6->src_address.as_u64[1];
      t->session_index = ~0;
    }
  return next_index;
}

#include <vnet/pipeline.h>

VLIB_NODE_FN (l2t_decap_node) (vlib_main_t * vm,
			       vlib_node_runtime_t * node,
			       vlib_frame_t * frame)
{
  return dispatch_pipeline (vm, node, frame);
}

/*
 * l2tp-decap and l2tp-decap-local have very slightly different behavior.
 * When a packet has no associated session l2tp-decap let it go to ip6 forward,
 * while l2tp-decap-local drops it.
 */

/* *INDENT-OFF* */
VLIB_REGISTER_NODE (l2t_decap_node) = {
  .name = "l2tp-decap",
  .vector_size = sizeof (u32),
  .format_trace = format_l2t_trace,
  .type = VLIB_NODE_TYPE_INTERNAL,

  .n_errors = ARRAY_LEN(l2t_decap_error_strings),
  .error_strings = l2t_decap_error_strings,

  .n_next_nodes = L2T_DECAP_N_NEXT,

  /* edit / add dispositions here */
  .next_nodes = {
        [L2T_DECAP_NEXT_L2_INPUT] = "l2-input",
        [L2T_DECAP_NEXT_DROP] = "error-drop",
  },
};
/* *INDENT-ON* */

extern vlib_node_function_t l2t_decap_node_fn;

/* *INDENT-OFF* */
VLIB_REGISTER_NODE (l2t_decap_local_node) = {
  .function = l2t_decap_node_fn,
  .name = "l2tp-decap-local",
  .vector_size = sizeof (u32),
  .format_trace = format_l2t_trace,
  .type = VLIB_NODE_TYPE_INTERNAL,

  .n_errors = ARRAY_LEN(l2t_decap_error_strings),
  .error_strings = l2t_decap_error_strings,

  .n_next_nodes = L2T_DECAP_N_NEXT,

  /* edit / add dispositions here */
  .next_nodes = {
    [L2T_DECAP_NEXT_L2_INPUT] = "l2-input",
    [L2T_DECAP_NEXT_DROP] = "error-drop",
  },
};
/* *INDENT-ON* */

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