LISP GPE: initial CP commit and DP improvements

Control Plane
-------------
In essence, this introduces basic support for map-request/reply
processing, the logic to generate and consume such messages, including
SMRs, a control-plane backend, consisting of an eid-table, locator and
locator-set tables, and CLI to interact with it. Naturally, we can now
serialize/deserialize LISP specific types: addresses, locators,
mappings, messages. An important caveat is that IPv6 support is not
complete, both for EIDs and RLOCs.

Functionally, the DP forwards all packets it can't handle to the CP
(lisp_cp_lookup node) which takes care of obtaining a mapping for the
packet's destination from a pre-configured map-resolver using the LISP
protocol. The CP then caches this information and programs the DP such
that all new packets with the same destination (or within the covering
prefix) are encapsulated to one of the locators retrieved in the
mapping. Ingress traffic-engineering is not yet supported.

Data Plane
----------
First of all, to enable punting to the CP, when LISP GPE is turned on a
default route that points to lisp_cp_lookup is now inserted. The DP
also exposes an API the CP can use to program forwarding for a given
mapping. This mainly consists in allocating a tunnel and programming the
FIB such that all packets destined to the mapping's prefix are forwarded
to a lisp-gpe encapsulating node.

Another important change done for lisp forwarding is that both source
and destination IP addresses are considered when encapsulating a packet.
To this end, a new FIB/mtrie is introduced as a second stage, src
lookup, post dst lookup. The latter is still done in the IP FIB but for
source-dest entries, in the dest adjacency the lookup_next_index points
to a lisp lookup node and the rewrite_header.sw_if_index points to the
src FIB. This is read by the lisp lookup node which then walks the src
mtrie, finds the associated adjacency, marks the buffer with the index
and forwards the packet to the appropriate next node (typically,
lisp-gpe-encap).

Change-Id: Ibdf52fdc1f89311854621403ccdd66f90e2522fd
Signed-off-by: Florin Coras <fcoras@cisco.com>

1 files changed, 126 insertions, 193 deletions

diff --git a/vnet/vnet/lisp-gpe/encap.c b/vnet/vnet/lisp-gpe/encap.c
index b3a52c464be..a8158782d8c 100644
--- a/vnet/vnet/lisp-gpe/encap.c
+++ b/vnet/vnet/lisp-gpe/encap.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2015 Cisco and/or its affiliates.
+ * Copyright (c) 2016 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:
@@ -12,10 +12,12 @@
  * 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/ip/udp.h>
 #include <vnet/ethernet/ethernet.h>
 #include <vnet/lisp-gpe/lisp_gpe.h>
 
@@ -31,45 +33,41 @@ static char * lisp_gpe_encap_error_strings[] = {
 
 typedef enum {
 #define _(sym,str) LISP_GPE_ENCAP_ERROR_##sym,
-    foreach_lisp_gpe_encap_error
+  foreach_lisp_gpe_encap_error
 #undef _
-    LISP_GPE_ENCAP_N_ERROR,
+  LISP_GPE_ENCAP_N_ERROR,
 } lisp_gpe_encap_error_t;
 
-typedef enum {
-    LISP_GPE_ENCAP_NEXT_IP4_LOOKUP,
-    LISP_GPE_ENCAP_NEXT_DROP,
-    LISP_GPE_ENCAP_N_NEXT,
+typedef enum
+{
+  LISP_GPE_ENCAP_NEXT_DROP,
+  LISP_GPE_ENCAP_NEXT_IP4_LOOKUP,
+  LISP_GPE_ENCAP_N_NEXT,
 } lisp_gpe_encap_next_t;
 
-typedef struct {
+typedef struct
+{
   u32 tunnel_index;
 } lisp_gpe_encap_trace_t;
 
-u8 * format_lisp_gpe_encap_trace (u8 * s, va_list * args)
+u8 *
+format_lisp_gpe_encap_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 *);
-  lisp_gpe_encap_trace_t * t 
-      = va_arg (*args, lisp_gpe_encap_trace_t *);
+  lisp_gpe_encap_trace_t * t = va_arg (*args, lisp_gpe_encap_trace_t *);
 
   s = format (s, "LISP-GPE-ENCAP: tunnel %d", t->tunnel_index);
   return s;
 }
 
-#define foreach_fixed_header_offset             \
-_(0) _(1) _(2) _(3) 
-
 static uword
-lisp_gpe_encap (vlib_main_t * vm,
-               vlib_node_runtime_t * node,
-               vlib_frame_t * from_frame)
+lisp_gpe_encap (vlib_main_t * vm, vlib_node_runtime_t * node,
+                vlib_frame_t * from_frame)
 {
   u32 n_left_from, next_index, * from, * to_next;
-  lisp_gpe_main_t * ngm = &lisp_gpe_main;
-  vnet_main_t * vnm = ngm->vnet_main;
+  lisp_gpe_main_t * lgm = &lisp_gpe_main;
   u32 pkts_encapsulated = 0;
-  u16 old_l0 = 0;
 
   from = vlib_frame_vector_args (from_frame);
   n_left_from = from_frame->n_vectors;
@@ -81,196 +79,131 @@ lisp_gpe_encap (vlib_main_t * vm,
       u32 n_left_to_next;
 
       vlib_get_next_frame (vm, node, next_index,
-			   to_next, n_left_to_next);
+                           to_next, n_left_to_next);
 
-#if 0
       while (n_left_from >= 4 && n_left_to_next >= 2)
-	{
-	  u32 bi0, bi1;
-	  vlib_buffer_t * b0, * b1;
-	  nsh_unicast_header_t * h0, * h1;
-          u32 label0, label1;
-	  u32 next0, next1;
-	  uword * p0, * p1;
-
-	  /* 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, 2*CLIB_CACHE_LINE_BYTES, LOAD);
-	    CLIB_PREFETCH (p3->data, 2*CLIB_CACHE_LINE_BYTES, LOAD);
-	  }
-
-	  bi0 = from[0];
-	  bi1 = from[1];
-	  to_next[0] = bi0;
-	  to_next[1] = bi1;
-	  from += 2;
-	  to_next += 2;
-	  n_left_to_next -= 2;
-	  n_left_from -= 2;
-
-	  b0 = vlib_get_buffer (vm, bi0);
-	  b1 = vlib_get_buffer (vm, bi1);
-
-          h0 = vlib_buffer_get_current (b0);
-          h1 = vlib_buffer_get_current (b1);
-          
-          next0 = next1 = NSH_INPUT_NEXT_IP4_INPUT;
-
-          label0 = clib_net_to_host_u32 (h0->label_exp_s_ttl);
-          label1 = clib_net_to_host_u32 (h1->label_exp_s_ttl);
-
-	  /* 
-	   * Translate label contents into a fib index.
-	   * This is a decent sanity check, and guarantees
-	   * a sane FIB for the downstream lookup
-	   */
-          label0 = vnet_nsh_uc_get_label (label0);
-          label1 = vnet_nsh_uc_get_label (label1);
-
-          /* If 2xlabels match, and match the 1-wide cache, use it */
-          if (label0 == label1 && rt->last_label == label0)
-            {
-              vnet_buffer(b0)->sw_if_index[VLIB_TX] = rt->last_fib_index;
-              vnet_buffer(b1)->sw_if_index[VLIB_TX] = rt->last_fib_index;
-            }
-          else
+        {
+          u32 bi0, bi1;
+          vlib_buffer_t * b0, * b1;
+          u32 next0, next1;
+          u32 adj_index0, adj_index1, tunnel_index0, tunnel_index1;
+          ip_adjacency_t * adj0, * adj1;
+          lisp_gpe_tunnel_t * t0, * t1;
+
+          next0 = next1 = LISP_GPE_ENCAP_NEXT_IP4_LOOKUP;
+
+          /* Prefetch next iteration. */
             {
-              p0 = hash_get (rt->mm->fib_index_by_nsh_label, label0);
-              if (PREDICT_FALSE (p0 == 0))
-                {
-                  next0 = NSH_INPUT_NEXT_DROP;
-                  b0->error = node->errors[NSH_ERROR_BAD_LABEL];
-                }
-              else
-                vnet_buffer(b0)->sw_if_index[VLIB_TX] = p0[0];
-              
-              p1 = hash_get (rt->mm->fib_index_by_nsh_label, label1);
-              if (PREDICT_FALSE (p1 == 0))
-                {
-                  next1 = NSH_INPUT_NEXT_DROP;
-                  b1->error = node->errors[NSH_ERROR_BAD_LABEL];
-                }
-              else
-                {
-                  vnet_buffer(b1)->sw_if_index[VLIB_TX] = p1[0];
-                  rt->last_fib_index = p1[0];
-                  rt->last_label = label1;
-                }
+              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, 2*CLIB_CACHE_LINE_BYTES, LOAD);
+              CLIB_PREFETCH(p3->data, 2*CLIB_CACHE_LINE_BYTES, LOAD);
             }
 
-          if (PREDICT_FALSE(b0->flags & VLIB_BUFFER_IS_TRACED)) 
+          bi0 = from[0];
+          bi1 = from[1];
+          to_next[0] = bi0;
+          to_next[1] = bi1;
+          from += 2;
+          to_next += 2;
+          n_left_to_next -= 2;
+          n_left_from -= 2;
+
+          b0 = vlib_get_buffer (vm, bi0);
+          b1 = vlib_get_buffer (vm, bi1);
+
+          /* Get adjacency and from it the tunnel_index */
+          adj_index0 = vnet_buffer(b0)->ip.adj_index[VLIB_TX];
+          adj_index1 = vnet_buffer(b1)->ip.adj_index[VLIB_TX];
+
+          adj0 = ip_get_adjacency (lgm->lookup_main, adj_index0);
+          adj1 = ip_get_adjacency (lgm->lookup_main, adj_index1);
+
+          tunnel_index0 = adj0->rewrite_header.sw_if_index;
+          tunnel_index1 = adj1->rewrite_header.sw_if_index;
+
+          t0 = pool_elt_at_index (lgm->tunnels, tunnel_index0);
+          t1 = pool_elt_at_index (lgm->tunnels, tunnel_index1);
+
+          ASSERT(t0 != 0);
+          ASSERT(t1 != 0);
+
+          ASSERT (sizeof(ip4_udp_lisp_gpe_header_t) == 36);
+          ip4_udp_encap_two (vm, b0, b1, t0->rewrite, t1->rewrite, 36);
+
+          /* Reset to look up tunnel partner in the configured FIB */
+          vnet_buffer(b0)->sw_if_index[VLIB_TX] = t0->encap_fib_index;
+          vnet_buffer(b1)->sw_if_index[VLIB_TX] = t1->encap_fib_index;
+
+          if (PREDICT_FALSE(b0->flags & VLIB_BUFFER_IS_TRACED))
             {
-              nsh_rx_trace_t *tr = vlib_add_trace (vm, node, 
-                                                   b0, sizeof (*tr));
-              tr->label_exp_s_ttl = label0;
+              lisp_gpe_encap_trace_t *tr = vlib_add_trace (vm, node, b0,
+                                                           sizeof(*tr));
+              tr->tunnel_index = t0 - lgm->tunnels;
             }
-          if (PREDICT_FALSE(b1->flags & VLIB_BUFFER_IS_TRACED)) 
+          if (PREDICT_FALSE(b1->flags & VLIB_BUFFER_IS_TRACED))
             {
-              nsh_rx_trace_t *tr = vlib_add_trace (vm, node, 
-                                                   b1, sizeof (*tr));
-              tr->label_exp_s_ttl = label1;
+              lisp_gpe_encap_trace_t *tr = vlib_add_trace (vm, node, b1,
+                                                           sizeof(*tr));
+              tr->tunnel_index = t1 - lgm->tunnels;
             }
 
-          vlib_buffer_advance (b0, sizeof (*h0));
-          vlib_buffer_advance (b1, sizeof (*h1));
+          pkts_encapsulated += 2;
+
+          vlib_validate_buffer_enqueue_x2(vm, node, next_index, to_next,
+                                          n_left_to_next, bi0, bi1, next0,
+                                          next1);
+        }
 
-	  vlib_validate_buffer_enqueue_x2 (vm, node, next_index,
-					   to_next, n_left_to_next,
-					   bi0, bi1, next0, next1);
-	}
-#endif 
-    
       while (n_left_from > 0 && n_left_to_next > 0)
-	{
-	  u32 bi0;
-	  vlib_buffer_t * b0;
-	  u32 next0 = LISP_GPE_ENCAP_NEXT_IP4_LOOKUP;
-          vnet_hw_interface_t * hi0;
-          ip4_header_t * ip0;
-          udp_header_t * udp0;
-          u64 * copy_src0, * copy_dst0;
-          u32 * copy_src_last0, * copy_dst_last0;
-          lisp_gpe_tunnel_t * t0;
-          u16 new_l0;
-          ip_csum_t sum0;
-
-	  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);
-
-          /* 1-wide cache? */
-          hi0 = vnet_get_sup_hw_interface 
-            (vnm, vnet_buffer(b0)->sw_if_index[VLIB_TX]);
-
-          t0 = pool_elt_at_index (ngm->tunnels, hi0->dev_instance);
-
-          ASSERT(vec_len(t0->rewrite) >= 24);
-
-          /* Apply the rewrite string. $$$$ vnet_rewrite? */
-          vlib_buffer_advance (b0, -(word)_vec_len(t0->rewrite));
-
-          ip0 = vlib_buffer_get_current(b0);
-          /* Copy the fixed header */
-          copy_dst0 = (u64 *) ip0;
-          copy_src0 = (u64 *) t0->rewrite;
-
-          ASSERT (sizeof (ip4_udp_lisp_gpe_header_t) == 36);
-
-          /* Copy first 32 octets 8-bytes at a time */
-#define _(offs) copy_dst0[offs] = copy_src0[offs];
-          foreach_fixed_header_offset;
-#undef _
-          /* Last 4 octets. Hopefully gcc will be our friend */
-          copy_dst_last0 = (u32 *)(&copy_dst0[4]);
-          copy_src_last0 = (u32 *)(&copy_src0[4]);
-          
-          copy_dst_last0[0] = copy_src_last0[0];
-
-          /* fix the <bleep>ing outer-IP checksum */
-          sum0 = ip0->checksum;
-          /* old_l0 always 0, see the rewrite setup */
-          new_l0 = 
-            clib_host_to_net_u16 (vlib_buffer_length_in_chain (vm, b0));
-          
-          sum0 = ip_csum_update (sum0, old_l0, new_l0, ip4_header_t,
-                                 length /* changed member */);
-          ip0->checksum = ip_csum_fold (sum0);
-          ip0->length = new_l0;
-          
-          /* Fix UDP length */
-          udp0 = (udp_header_t *)(ip0+1);
-          new_l0 = clib_host_to_net_u16 (vlib_buffer_length_in_chain (vm, b0)
-                                         - sizeof (*ip0));
-          
-          udp0->length = new_l0;
+        {
+          vlib_buffer_t * b0;
+          u32 bi0, adj_index0, tunnel_index0;
+          u32 next0 = LISP_GPE_ENCAP_NEXT_IP4_LOOKUP;
+          lisp_gpe_tunnel_t * t0 = 0;
+          ip_adjacency_t * adj0;
+
+          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);
+
+          /* Get adjacency and from it the tunnel_index */
+          adj_index0 = vnet_buffer(b0)->ip.adj_index[VLIB_TX];
+          adj0 = ip_get_adjacency (lgm->lookup_main, adj_index0);
+
+          tunnel_index0 = adj0->rewrite_header.sw_if_index;
+          t0 = pool_elt_at_index (lgm->tunnels, tunnel_index0);
+
+          ASSERT(t0 != 0);
+
+          ASSERT (sizeof(ip4_udp_lisp_gpe_header_t) == 36);
+          ip4_udp_encap_one (vm, b0, t0->rewrite, 36);
 
           /* Reset to look up tunnel partner in the configured FIB */
           vnet_buffer(b0)->sw_if_index[VLIB_TX] = t0->encap_fib_index;
-          pkts_encapsulated ++;
+
+          pkts_encapsulated++;
 
           if (PREDICT_FALSE(b0->flags & VLIB_BUFFER_IS_TRACED)) 
             {
-              lisp_gpe_encap_trace_t *tr = 
-                vlib_add_trace (vm, node, b0, sizeof (*tr));
-              tr->tunnel_index = t0 - ngm->tunnels;
+              lisp_gpe_encap_trace_t *tr = vlib_add_trace (vm, node, b0,
+                                                           sizeof(*tr));
+              tr->tunnel_index = t0 - lgm->tunnels;
             }
-	  vlib_validate_buffer_enqueue_x1 (vm, node, next_index,
-					   to_next, n_left_to_next,
-					   bi0, next0);
-	}
+          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);
     }
@@ -293,7 +226,7 @@ VLIB_REGISTER_NODE (lisp_gpe_encap_node) = {
   .n_next_nodes = LISP_GPE_ENCAP_N_NEXT,
 
   .next_nodes = {
-        [LISP_GPE_ENCAP_NEXT_IP4_LOOKUP] = "ip4-lookup",
-        [LISP_GPE_ENCAP_NEXT_DROP] = "error-drop",
+      [LISP_GPE_ENCAP_NEXT_DROP] = "error-drop",
+      [LISP_GPE_ENCAP_NEXT_IP4_LOOKUP] = "ip4-lookup",
   },
 };