bond: Add bonding driver and LACP protocol

Add bonding driver to support creation of bond interface which composes of
multiple slave interfaces. The slave interfaces could be physical interfaces,
or just any virtual interfaces. For example, memif interfaces.

The syntax to create a bond interface is
create bond mode <lacp | xor | acitve-backup | broadcast | round-robin>

To enslave an interface to the bond interface,
enslave interface TenGigabitEthernet6/0/0 to BondEthernet0

Please see src/plugins/lacp/lacp_doc.md for more examples and additional
options.

LACP is a control plane protocol which manages and monitors the status of
the slave interfaces. The protocol is part of 802.3ad standard. This patch
implements LACPv1. LACPv2 is not supported.
To enable LACP on the bond interface, specify "mode lacp" when the bond
interface is created. The syntax to enslave a slave interface is the same as
other bonding modes.

Change-Id: I06581d3b87635972f9f0e1ec50b67560fc13e26c
Signed-off-by: Steven <sluong@cisco.com>

1 files changed, 359 insertions, 0 deletions

diff --git a/src/plugins/lacp/input.c b/src/plugins/lacp/input.c
new file mode 100644
index 00000000000..45db3b8455b
--- /dev/null
+++ b/src/plugins/lacp/input.c
@@ -0,0 +1,359 @@
+/*
+ * 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.
+ */
+
+#define _GNU_SOURCE
+#include <vnet/bonding/node.h>
+#include <lacp/node.h>
+
+static int
+lacp_packet_scan (vlib_main_t * vm, slave_if_t * sif)
+{
+  lacp_pdu_t *lacpdu = (lacp_pdu_t *) sif->last_rx_pkt;
+
+  if (lacpdu->subtype != LACP_SUBTYPE)
+    return LACP_ERROR_UNSUPPORTED;
+
+  /*
+   * According to the spec, no checking on the version number and tlv types.
+   * But we may check the tlv lengths.
+   */
+  if ((lacpdu->actor.tlv_length != sizeof (lacp_actor_partner_t)) ||
+      (lacpdu->partner.tlv_length != sizeof (lacp_actor_partner_t)) ||
+      (lacpdu->collector.tlv_length != sizeof (lacp_collector_t)) ||
+      (lacpdu->terminator.tlv_length != 0))
+    return (LACP_ERROR_BAD_TLV);
+
+  lacp_machine_dispatch (&lacp_rx_machine, vm, sif,
+			 LACP_RX_EVENT_PDU_RECEIVED, &sif->rx_state);
+
+  return LACP_ERROR_NONE;
+}
+
+static void
+marker_fill_pdu (marker_pdu_t * marker, slave_if_t * sif)
+{
+  marker_pdu_t *pkt = (marker_pdu_t *) sif->last_marker_pkt;
+
+  marker->marker_info = pkt->marker_info;
+  marker->marker_info.tlv_type = MARKER_RESPONSE_INFORMATION;
+}
+
+void
+marker_fill_request_pdu (marker_pdu_t * marker, slave_if_t * sif)
+{
+  marker->marker_info.tlv_type = MARKER_INFORMATION;
+  marker->marker_info.requester_port = sif->actor.port_number;
+  clib_memcpy (marker->marker_info.requester_system, sif->actor.system, 6);
+  marker->marker_info.requester_transaction_id = sif->marker_tx_id;
+  sif->marker_tx_id++;
+}
+
+static void
+send_ethernet_marker_response_pdu (slave_if_t * sif)
+{
+  lacp_main_t *lm = &lacp_main;
+  u32 *to_next;
+  ethernet_marker_pdu_t *h0;
+  vnet_hw_interface_t *hw;
+  u32 bi0;
+  vlib_buffer_t *b0;
+  vlib_frame_t *f;
+  vlib_main_t *vm = lm->vlib_main;
+  vnet_main_t *vnm = lm->vnet_main;
+
+  /*
+   * see lacp_periodic_init() to understand what's already painted
+   * into the buffer by the packet template mechanism
+   */
+  h0 = vlib_packet_template_get_packet
+    (vm, &lm->marker_packet_templates[sif->packet_template_index], &bi0);
+
+  if (!h0)
+    return;
+
+  /* Add the interface's ethernet source address */
+  hw = vnet_get_sup_hw_interface (vnm, sif->sw_if_index);
+
+  clib_memcpy (h0->ethernet.src_address, hw->hw_address,
+	       vec_len (hw->hw_address));
+
+  marker_fill_pdu (&h0->marker, sif);
+
+  /* Set the outbound packet length */
+  b0 = vlib_get_buffer (vm, bi0);
+  b0->current_length = sizeof (ethernet_marker_pdu_t);
+  b0->current_data = 0;
+  b0->total_length_not_including_first_buffer = 0;
+
+  /* And the outbound interface */
+  vnet_buffer (b0)->sw_if_index[VLIB_TX] = hw->sw_if_index;
+
+  /* And output the packet on the correct interface */
+  f = vlib_get_frame_to_node (vm, hw->output_node_index);
+
+  to_next = vlib_frame_vector_args (f);
+  to_next[0] = bi0;
+  f->n_vectors = 1;
+
+  vlib_put_frame_to_node (vm, hw->output_node_index, f);
+}
+
+static int
+handle_marker_protocol (vlib_main_t * vm, slave_if_t * sif)
+{
+  marker_pdu_t *marker = (marker_pdu_t *) sif->last_marker_pkt;
+
+  /*
+   * According to the spec, no checking on the version number and tlv types.
+   * But we may check the tlv lengths.
+   */
+  if ((marker->marker_info.tlv_length != sizeof (marker_information_t)) ||
+      (marker->terminator.tlv_length != 0))
+    return (LACP_ERROR_BAD_TLV);
+
+  send_ethernet_marker_response_pdu (sif);
+
+  return LACP_ERROR_NONE;
+}
+
+/*
+ * lacp input routine
+ */
+lacp_error_t
+lacp_input (vlib_main_t * vm, vlib_buffer_t * b0, u32 bi0)
+{
+  lacp_main_t *lm = &lacp_main;
+  slave_if_t *sif;
+  uword nbytes;
+  lacp_error_t e;
+  marker_pdu_t *marker;
+  uword last_packet_signature;
+  bond_if_t *bif;
+
+  sif =
+    bond_get_slave_by_sw_if_index (vnet_buffer (b0)->sw_if_index[VLIB_RX]);
+  if ((sif == 0) || (sif->mode != BOND_MODE_LACP))
+    {
+      return LACP_ERROR_DISABLED;
+    }
+
+  /* Handle marker protocol */
+  marker = (marker_pdu_t *) (b0->data + b0->current_data);
+  if (marker->subtype == MARKER_SUBTYPE)
+    {
+      if (sif->last_marker_pkt)
+	_vec_len (sif->last_marker_pkt) = 0;
+      vec_validate (sif->last_marker_pkt,
+		    vlib_buffer_length_in_chain (vm, b0) - 1);
+      nbytes = vlib_buffer_contents (vm, bi0, sif->last_marker_pkt);
+      ASSERT (nbytes <= vec_len (sif->last_marker_pkt));
+      if (nbytes < sizeof (lacp_pdu_t))
+	return LACP_ERROR_TOO_SMALL;
+      return (handle_marker_protocol (vm, sif));
+    }
+
+  /*
+   * typical clib idiom. Don't repeatedly allocate and free
+   * the per-neighbor rx buffer. Reset its apparent length to zero
+   * and reuse it.
+   */
+  if (sif->last_rx_pkt)
+    _vec_len (sif->last_rx_pkt) = 0;
+
+  /*
+   * Make sure the per-neighbor rx buffer is big enough to hold
+   * the data we're about to copy
+   */
+  vec_validate (sif->last_rx_pkt, vlib_buffer_length_in_chain (vm, b0) - 1);
+
+  /*
+   * Coalesce / copy the buffer chain into the per-neighbor
+   * rx buffer
+   */
+  nbytes = vlib_buffer_contents (vm, bi0, sif->last_rx_pkt);
+  ASSERT (nbytes <= vec_len (sif->last_rx_pkt));
+
+  if (nbytes < sizeof (lacp_pdu_t))
+    {
+      return LACP_ERROR_TOO_SMALL;
+    }
+
+  last_packet_signature =
+    hash_memory (sif->last_rx_pkt, vec_len (sif->last_rx_pkt), 0xd00b);
+
+  bif = bond_get_master_by_dev_instance (sif->bif_dev_instance);
+  if (sif->last_packet_signature_valid &&
+      (sif->last_packet_signature == last_packet_signature) &&
+      hash_get (bif->active_slave_by_sw_if_index, sif->sw_if_index))
+    {
+      lacp_start_current_while_timer (lm->vlib_main, sif,
+				      sif->ttl_in_seconds);
+      e = LACP_ERROR_CACHE_HIT;
+    }
+  else
+    {
+      /* Actually scan the packet */
+      e = lacp_packet_scan (vm, sif);
+      sif->last_packet_signature_valid = 1;
+      sif->last_packet_signature = last_packet_signature;
+    }
+
+  if (sif->last_rx_pkt)
+    _vec_len (sif->last_rx_pkt) = 0;
+
+  return e;
+}
+
+/*
+ * setup neighbor hash table
+ */
+static clib_error_t *
+lacp_init (vlib_main_t * vm)
+{
+  clib_error_t *error;
+
+  if ((error = vlib_call_init_function (vm, lacp_periodic_init)))
+    return error;
+
+  return 0;
+}
+
+VLIB_INIT_FUNCTION (lacp_init);
+
+/*
+ * packet trace format function, very similar to
+ * lacp_packet_scan except that we call the per TLV format
+ * functions instead of the per TLV processing functions
+ */
+u8 *
+lacp_input_format_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 *);
+  lacp_input_trace_t *t = va_arg (*args, lacp_input_trace_t *);
+  lacp_pdu_t *lacpdu = &t->pkt.lacpdu;
+  marker_pdu_t *marker = &t->pkt.marker;
+  int i, len;
+  u8 *p;
+  lacp_state_struct *state_entry;
+
+  s = format (s, "Length: %d\n", t->len);
+  if (t->len >= sizeof (lacp_pdu_t))
+    {
+      switch (lacpdu->subtype)
+	{
+	case MARKER_SUBTYPE:
+	  if (marker->version_number == MARKER_PROTOCOL_VERSION)
+	    s = format (s, "  Markerv1\n");
+	  else
+	    s = format (s, "  Subtype %u, Version %u\n", marker->subtype,
+			marker->version_number);
+	  s = format (s, "  Marker Information TLV: type %u\n",
+		      marker->marker_info.tlv_type);
+	  s = format (s, "  Marker Information TLV: length %u\n",
+		      marker->marker_info.tlv_length);
+	  s = format (s, "  Requester port: %u\n",
+		      marker->marker_info.requester_port);
+	  s = format (s, "  Requester system: %U\n", format_ethernet_address,
+		      marker->marker_info.requester_system);
+	  s = format (s, "  Requester transaction ID: %u\n",
+		      marker->marker_info.requester_transaction_id);
+	  break;
+
+	case LACP_SUBTYPE:
+	  if (lacpdu->version_number == LACP_ACTOR_LACP_VERSION)
+	    s = format (s, "  LACPv1\n");
+	  else
+	    s = format (s, "  Subtype %u, Version %u\n", lacpdu->subtype,
+			lacpdu->version_number);
+	  s = format (s, "  Actor Information TLV: length %u\n",
+		      lacpdu->actor.tlv_length);
+	  s = format (s, "    System %U\n", format_ethernet_address,
+		      lacpdu->actor.port_info.system);
+	  s = format (s, "    System priority %u\n",
+		      ntohs (lacpdu->actor.port_info.system_priority));
+	  s = format (s, "    Key %u\n", ntohs (lacpdu->actor.port_info.key));
+	  s = format (s, "    Port priority %u\n",
+		      ntohs (lacpdu->actor.port_info.port_priority));
+	  s = format (s, "    Port number %u\n",
+		      ntohs (lacpdu->actor.port_info.port_number));
+	  s = format (s, "    State 0x%x\n", lacpdu->actor.port_info.state);
+	  state_entry = (lacp_state_struct *) & lacp_state_array;
+	  while (state_entry->str)
+	    {
+	      if (lacpdu->actor.port_info.state & (1 << state_entry->bit))
+		s = format (s, "      %s (%d)\n", state_entry->str,
+			    state_entry->bit);
+	      state_entry++;
+	    }
+
+	  s = format (s, "  Partner Information TLV: length %u\n",
+		      lacpdu->partner.tlv_length);
+	  s = format (s, "    System %U\n", format_ethernet_address,
+		      lacpdu->partner.port_info.system);
+	  s = format (s, "    System priority %u\n",
+		      ntohs (lacpdu->partner.port_info.system_priority));
+	  s =
+	    format (s, "    Key %u\n", ntohs (lacpdu->partner.port_info.key));
+	  s =
+	    format (s, "    Port priority %u\n",
+		    ntohs (lacpdu->partner.port_info.port_priority));
+	  s =
+	    format (s, "    Port number %u\n",
+		    ntohs (lacpdu->partner.port_info.port_number));
+	  s = format (s, "    State 0x%x\n", lacpdu->partner.port_info.state);
+	  state_entry = (lacp_state_struct *) & lacp_state_array;
+	  while (state_entry->str)
+	    {
+	      if (lacpdu->partner.port_info.state & (1 << state_entry->bit))
+		s = format (s, "      %s (%d)\n", state_entry->str,
+			    state_entry->bit);
+	      state_entry++;
+	    }
+	  break;
+
+	default:
+	  break;
+	}
+    }
+
+  if (t->len > sizeof (lacp_pdu_t))
+    len = sizeof (lacp_pdu_t);
+  else
+    len = t->len;
+  p = (u8 *) lacpdu;
+  for (i = 0; i < len; i++)
+    {
+      if ((i % 16) == 0)
+	{
+	  if (i)
+	    s = format (s, "\n");
+	  s = format (s, "  0x%04x: ", i);
+	}
+      if ((i % 2) == 0)
+	s = format (s, " ");
+      s = format (s, "%02x", p[i]);
+    }
+
+  return s;
+}
+
+/*
+ * fd.io coding-style-patch-verification: ON
+ *
+ * Local Variables:
+ * eval: (c-set-style "gnu")
+ * End:
+ */