summaryrefslogtreecommitdiffstats
path: root/src/vnet/l2/l2_input.c
diff options
context:
space:
mode:
authorDamjan Marion <damarion@cisco.com>2016-12-19 23:05:39 +0100
committerDamjan Marion <damarion@cisco.com>2016-12-28 12:25:14 +0100
commit7cd468a3d7dee7d6c92f69a0bb7061ae208ec727 (patch)
tree5de62f8dbd3a752f5a676ca600e43d2652d1ff1a /src/vnet/l2/l2_input.c
parent696f1adec0df3b8f161862566dd9c86174302658 (diff)
Reorganize source tree to use single autotools instance
Change-Id: I7b51f88292e057c6443b12224486f2d0c9f8ae23 Signed-off-by: Damjan Marion <damarion@cisco.com>
Diffstat (limited to 'src/vnet/l2/l2_input.c')
-rw-r--r--src/vnet/l2/l2_input.c1116
1 files changed, 1116 insertions, 0 deletions
diff --git a/src/vnet/l2/l2_input.c b/src/vnet/l2/l2_input.c
new file mode 100644
index 00000000000..a104ec9eebb
--- /dev/null
+++ b/src/vnet/l2/l2_input.c
@@ -0,0 +1,1116 @@
+/*
+ * l2_input.c : layer 2 input packet processing
+ *
+ * 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 <vlib/vlib.h>
+#include <vnet/vnet.h>
+#include <vnet/pg/pg.h>
+#include <vnet/ethernet/ethernet.h>
+#include <vnet/ethernet/packet.h>
+#include <vnet/ip/ip_packet.h>
+#include <vnet/ip/ip4_packet.h>
+#include <vnet/ip/ip6_packet.h>
+#include <vlib/cli.h>
+#include <vnet/l2/l2_input.h>
+#include <vnet/l2/l2_output.h>
+#include <vnet/l2/feat_bitmap.h>
+#include <vnet/l2/l2_bvi.h>
+#include <vnet/l2/l2_fib.h>
+
+#include <vppinfra/error.h>
+#include <vppinfra/hash.h>
+#include <vppinfra/cache.h>
+
+/**
+ * @file
+ * @brief Interface Input Mode (Layer 2 Cross-Connect or Bridge / Layer 3).
+ *
+ * This file contains the CLI Commands that modify the input mode of an
+ * interface. For interfaces in a Layer 2 cross-connect, all packets
+ * received on one interface will be transmitted to the other. For
+ * interfaces in a bridge-domain, packets will be forwarded to other
+ * interfaces in the same bridge-domain based on destination mac address.
+ * For interfaces in Layer 3 mode, the packets will be routed.
+ */
+
+/* Feature graph node names */
+static char *l2input_feat_names[] = {
+#define _(sym,name) name,
+ foreach_l2input_feat
+#undef _
+};
+
+char **
+l2input_get_feat_names (void)
+{
+ return l2input_feat_names;
+}
+
+
+typedef struct
+{
+ /* per-pkt trace data */
+ u8 src[6];
+ u8 dst[6];
+ u32 next_index;
+ u32 sw_if_index;
+} l2input_trace_t;
+
+/* packet trace format function */
+static u8 *
+format_l2input_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 *);
+ l2input_trace_t *t = va_arg (*args, l2input_trace_t *);
+
+ s = format (s, "l2-input: sw_if_index %d dst %U src %U",
+ t->sw_if_index,
+ format_ethernet_address, t->dst,
+ format_ethernet_address, t->src);
+ return s;
+}
+
+l2input_main_t l2input_main;
+
+#define foreach_l2input_error \
+_(L2INPUT, "L2 input packets") \
+_(DROP, "L2 input drops")
+
+typedef enum
+{
+#define _(sym,str) L2INPUT_ERROR_##sym,
+ foreach_l2input_error
+#undef _
+ L2INPUT_N_ERROR,
+} l2input_error_t;
+
+static char *l2input_error_strings[] = {
+#define _(sym,string) string,
+ foreach_l2input_error
+#undef _
+};
+
+typedef enum
+{ /* */
+ L2INPUT_NEXT_LEARN,
+ L2INPUT_NEXT_FWD,
+ L2INPUT_NEXT_DROP,
+ L2INPUT_N_NEXT,
+} l2input_next_t;
+
+
+static_always_inline void
+classify_and_dispatch (vlib_main_t * vm,
+ vlib_node_runtime_t * node,
+ u32 cpu_index,
+ l2input_main_t * msm, vlib_buffer_t * b0, u32 * next0)
+{
+ /*
+ * Load L2 input feature struct
+ * Load bridge domain struct
+ * Parse ethernet header to determine unicast/mcast/broadcast
+ * take L2 input stat
+ * classify packet as IP/UDP/TCP, control, other
+ * mask feature bitmap
+ * go to first node in bitmap
+ * Later: optimize VTM
+ *
+ * For L2XC,
+ * set tx sw-if-handle
+ */
+
+ u16 ethertype;
+ u8 protocol;
+ l2_input_config_t *config;
+ l2_bridge_domain_t *bd_config;
+ u16 bd_index0;
+ u32 feature_bitmap;
+ u32 feat_mask;
+ ethernet_header_t *h0;
+ u8 *l3h0;
+ u32 sw_if_index0;
+
+#define get_u16(addr) ( *((u16 *)(addr)) )
+
+ sw_if_index0 = vnet_buffer (b0)->sw_if_index[VLIB_RX];
+
+ h0 = vlib_buffer_get_current (b0);
+ l3h0 = (u8 *) h0 + vnet_buffer (b0)->l2.l2_len;
+
+ ethertype = clib_net_to_host_u16 (get_u16 (l3h0 - 2));
+ feat_mask = ~0;
+
+ /* Get config for the input interface */
+ config = vec_elt_at_index (msm->configs, sw_if_index0);
+
+ /* Save split horizon group */
+ vnet_buffer (b0)->l2.shg = config->shg;
+
+ /* determine layer2 kind for stat and mask */
+ if (PREDICT_FALSE (ethernet_address_cast (h0->dst_address)))
+ {
+ protocol = ((ip6_header_t *) l3h0)->protocol;
+
+ /* Disable bridge forwarding (flooding will execute instead if not xconnect) */
+ feat_mask &= ~(L2INPUT_FEAT_FWD | L2INPUT_FEAT_UU_FLOOD);
+
+ /* Disable ARP-term for non-ARP and non-ICMP6 packet */
+ if (ethertype != ETHERNET_TYPE_ARP &&
+ (ethertype != ETHERNET_TYPE_IP6 || protocol != IP_PROTOCOL_ICMP6))
+ feat_mask &= ~(L2INPUT_FEAT_ARP_TERM);
+ }
+ else
+ {
+ /*
+ * Check for from-BVI processing - set SHG of unicast packets from BVI
+ * to 0 so it is not dropped for VXLAN tunnels or other ports with the
+ * same SHG as that of the BVI.
+ */
+ if (PREDICT_FALSE (vnet_buffer (b0)->sw_if_index[VLIB_TX] ==
+ L2INPUT_BVI))
+ vnet_buffer (b0)->l2.shg = 0;
+ }
+
+
+ if (config->xconnect)
+ {
+ /* Set the output interface */
+ vnet_buffer (b0)->sw_if_index[VLIB_TX] = config->output_sw_if_index;
+ }
+ else
+ {
+ /* Do bridge-domain processing */
+ bd_index0 = config->bd_index;
+ /* save BD ID for next feature graph nodes */
+ vnet_buffer (b0)->l2.bd_index = bd_index0;
+
+ /* Get config for the bridge domain interface */
+ bd_config = vec_elt_at_index (msm->bd_configs, bd_index0);
+
+ /*
+ * Process bridge domain feature enables.
+ * To perform learning/flooding/forwarding, the corresponding bit
+ * must be enabled in both the input interface config and in the
+ * bridge domain config. In the bd_bitmap, bits for features other
+ * than learning/flooding/forwarding should always be set.
+ */
+ feat_mask = feat_mask & bd_config->feature_bitmap;
+ }
+
+ /* mask out features from bitmap using packet type and bd config */
+ feature_bitmap = config->feature_bitmap & feat_mask;
+
+ /* save for next feature graph nodes */
+ vnet_buffer (b0)->l2.feature_bitmap = feature_bitmap;
+
+ /* Determine the next node */
+ *next0 = feat_bitmap_get_next_node_index (msm->feat_next_node_index,
+ feature_bitmap);
+}
+
+
+static uword
+l2input_node_fn (vlib_main_t * vm,
+ vlib_node_runtime_t * node, vlib_frame_t * frame)
+{
+ u32 n_left_from, *from, *to_next;
+ l2input_next_t next_index;
+ l2input_main_t *msm = &l2input_main;
+ u32 cpu_index = os_get_cpu_number ();
+
+ from = vlib_frame_vector_args (frame);
+ n_left_from = frame->n_vectors; /* number of packets to process */
+ next_index = node->cached_next_index;
+
+ while (n_left_from > 0)
+ {
+ u32 n_left_to_next;
+
+ /* get space to enqueue frame to graph node "next_index" */
+ vlib_get_next_frame (vm, node, next_index, to_next, n_left_to_next);
+
+ while (n_left_from >= 8 && n_left_to_next >= 4)
+ {
+ u32 bi0, bi1, bi2, bi3;
+ vlib_buffer_t *b0, *b1, *b2, *b3;
+ u32 next0, next1, next2, next3;
+ u32 sw_if_index0, sw_if_index1, sw_if_index2, sw_if_index3;
+
+ /* Prefetch next iteration. */
+ {
+ vlib_buffer_t *p4, *p5, *p6, *p7;
+
+ p4 = vlib_get_buffer (vm, from[4]);
+ p5 = vlib_get_buffer (vm, from[5]);
+ p6 = vlib_get_buffer (vm, from[6]);
+ p7 = vlib_get_buffer (vm, from[7]);
+
+ /* Prefetch the buffer header and packet for the N+2 loop iteration */
+ vlib_prefetch_buffer_header (p4, LOAD);
+ vlib_prefetch_buffer_header (p5, LOAD);
+ vlib_prefetch_buffer_header (p6, LOAD);
+ vlib_prefetch_buffer_header (p7, LOAD);
+
+ CLIB_PREFETCH (p4->data, CLIB_CACHE_LINE_BYTES, STORE);
+ CLIB_PREFETCH (p5->data, CLIB_CACHE_LINE_BYTES, STORE);
+ CLIB_PREFETCH (p6->data, CLIB_CACHE_LINE_BYTES, STORE);
+ CLIB_PREFETCH (p7->data, CLIB_CACHE_LINE_BYTES, STORE);
+
+ /*
+ * Don't bother prefetching the bridge-domain config (which
+ * depends on the input config above). Only a small number of
+ * bridge domains are expected. Plus the structure is small
+ * and several fit in a cache line.
+ */
+ }
+
+ /* speculatively enqueue b0 and b1 to the current next frame */
+ /* bi is "buffer index", b is pointer to the buffer */
+ to_next[0] = bi0 = from[0];
+ to_next[1] = bi1 = from[1];
+ to_next[2] = bi2 = from[2];
+ to_next[3] = bi3 = from[3];
+ from += 4;
+ to_next += 4;
+ n_left_from -= 4;
+ n_left_to_next -= 4;
+
+ b0 = vlib_get_buffer (vm, bi0);
+ b1 = vlib_get_buffer (vm, bi1);
+ b2 = vlib_get_buffer (vm, bi2);
+ b3 = vlib_get_buffer (vm, bi3);
+
+ if (PREDICT_FALSE ((node->flags & VLIB_NODE_FLAG_TRACE)))
+ {
+ /* RX interface handles */
+ sw_if_index0 = vnet_buffer (b0)->sw_if_index[VLIB_RX];
+ sw_if_index1 = vnet_buffer (b1)->sw_if_index[VLIB_RX];
+ sw_if_index2 = vnet_buffer (b2)->sw_if_index[VLIB_RX];
+ sw_if_index3 = vnet_buffer (b3)->sw_if_index[VLIB_RX];
+
+ if (b0->flags & VLIB_BUFFER_IS_TRACED)
+ {
+ ethernet_header_t *h0 = vlib_buffer_get_current (b0);
+ l2input_trace_t *t =
+ vlib_add_trace (vm, node, b0, sizeof (*t));
+ t->sw_if_index = sw_if_index0;
+ clib_memcpy (t->src, h0->src_address, 6);
+ clib_memcpy (t->dst, h0->dst_address, 6);
+ }
+ if (b1->flags & VLIB_BUFFER_IS_TRACED)
+ {
+ ethernet_header_t *h1 = vlib_buffer_get_current (b1);
+ l2input_trace_t *t =
+ vlib_add_trace (vm, node, b1, sizeof (*t));
+ t->sw_if_index = sw_if_index1;
+ clib_memcpy (t->src, h1->src_address, 6);
+ clib_memcpy (t->dst, h1->dst_address, 6);
+ }
+ if (b2->flags & VLIB_BUFFER_IS_TRACED)
+ {
+ ethernet_header_t *h2 = vlib_buffer_get_current (b2);
+ l2input_trace_t *t =
+ vlib_add_trace (vm, node, b2, sizeof (*t));
+ t->sw_if_index = sw_if_index2;
+ clib_memcpy (t->src, h2->src_address, 6);
+ clib_memcpy (t->dst, h2->dst_address, 6);
+ }
+ if (b3->flags & VLIB_BUFFER_IS_TRACED)
+ {
+ ethernet_header_t *h3 = vlib_buffer_get_current (b3);
+ l2input_trace_t *t =
+ vlib_add_trace (vm, node, b3, sizeof (*t));
+ t->sw_if_index = sw_if_index3;
+ clib_memcpy (t->src, h3->src_address, 6);
+ clib_memcpy (t->dst, h3->dst_address, 6);
+ }
+ }
+
+ vlib_node_increment_counter (vm, l2input_node.index,
+ L2INPUT_ERROR_L2INPUT, 4);
+
+ classify_and_dispatch (vm, node, cpu_index, msm, b0, &next0);
+ classify_and_dispatch (vm, node, cpu_index, msm, b1, &next1);
+ classify_and_dispatch (vm, node, cpu_index, msm, b2, &next2);
+ classify_and_dispatch (vm, node, cpu_index, msm, b3, &next3);
+
+ /* verify speculative enqueues, maybe switch current next frame */
+ /* if next0==next1==next_index then nothing special needs to be done */
+ vlib_validate_buffer_enqueue_x4 (vm, node, next_index,
+ to_next, n_left_to_next,
+ bi0, bi1, bi2, bi3,
+ next0, next1, next2, next3);
+ }
+
+ while (n_left_from > 0 && n_left_to_next > 0)
+ {
+ u32 bi0;
+ vlib_buffer_t *b0;
+ u32 next0;
+ u32 sw_if_index0;
+
+ /* 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);
+
+ if (PREDICT_FALSE ((node->flags & VLIB_NODE_FLAG_TRACE)
+ && (b0->flags & VLIB_BUFFER_IS_TRACED)))
+ {
+ ethernet_header_t *h0 = vlib_buffer_get_current (b0);
+ l2input_trace_t *t = vlib_add_trace (vm, node, b0, sizeof (*t));
+ sw_if_index0 = vnet_buffer (b0)->sw_if_index[VLIB_RX];
+ t->sw_if_index = sw_if_index0;
+ clib_memcpy (t->src, h0->src_address, 6);
+ clib_memcpy (t->dst, h0->dst_address, 6);
+ }
+
+ vlib_node_increment_counter (vm, l2input_node.index,
+ L2INPUT_ERROR_L2INPUT, 1);
+
+ classify_and_dispatch (vm, node, cpu_index, msm, b0, &next0);
+
+ /* 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;
+}
+
+
+/* *INDENT-OFF* */
+VLIB_REGISTER_NODE (l2input_node) = {
+ .function = l2input_node_fn,
+ .name = "l2-input",
+ .vector_size = sizeof (u32),
+ .format_trace = format_l2input_trace,
+ .format_buffer = format_ethernet_header_with_length,
+ .type = VLIB_NODE_TYPE_INTERNAL,
+
+ .n_errors = ARRAY_LEN(l2input_error_strings),
+ .error_strings = l2input_error_strings,
+
+ .n_next_nodes = L2INPUT_N_NEXT,
+
+ /* edit / add dispositions here */
+ .next_nodes = {
+ [L2INPUT_NEXT_LEARN] = "l2-learn",
+ [L2INPUT_NEXT_FWD] = "l2-fwd",
+ [L2INPUT_NEXT_DROP] = "error-drop",
+ },
+};
+/* *INDENT-ON* */
+
+VLIB_NODE_FUNCTION_MULTIARCH (l2input_node, l2input_node_fn)
+ clib_error_t *l2input_init (vlib_main_t * vm)
+{
+ l2input_main_t *mp = &l2input_main;
+
+ mp->vlib_main = vm;
+ mp->vnet_main = vnet_get_main ();
+
+ /* Get packets RX'd from L2 interfaces */
+ ethernet_register_l2_input (vm, l2input_node.index);
+
+ /* Create the config vector */
+ vec_validate (mp->configs, 100);
+ /* create 100 sw interface entries and zero them */
+
+ /* Initialize the feature next-node indexes */
+ feat_bitmap_init_next_nodes (vm,
+ l2input_node.index,
+ L2INPUT_N_FEAT,
+ l2input_get_feat_names (),
+ mp->feat_next_node_index);
+
+ return 0;
+}
+
+VLIB_INIT_FUNCTION (l2input_init);
+
+
+/** Get a pointer to the config for the given interface. */
+l2_input_config_t *
+l2input_intf_config (u32 sw_if_index)
+{
+ l2input_main_t *mp = &l2input_main;
+
+ vec_validate (mp->configs, sw_if_index);
+ return vec_elt_at_index (mp->configs, sw_if_index);
+}
+
+/** Enable (or disable) the feature in the bitmap for the given interface. */
+u32
+l2input_intf_bitmap_enable (u32 sw_if_index, u32 feature_bitmap, u32 enable)
+{
+ l2input_main_t *mp = &l2input_main;
+ l2_input_config_t *config;
+
+ vec_validate (mp->configs, sw_if_index);
+ config = vec_elt_at_index (mp->configs, sw_if_index);
+
+ if (enable)
+ {
+ config->feature_bitmap |= feature_bitmap;
+ }
+ else
+ {
+ config->feature_bitmap &= ~feature_bitmap;
+ }
+
+ return config->feature_bitmap;
+}
+
+u32
+l2input_set_bridge_features (u32 bd_index, u32 feat_mask, u32 feat_value)
+{
+ l2_bridge_domain_t *bd_config;
+ vec_validate (l2input_main.bd_configs, bd_index);
+ bd_config = vec_elt_at_index (l2input_main.bd_configs, bd_index);
+ bd_validate (bd_config);
+ bd_config->feature_bitmap =
+ (bd_config->feature_bitmap & ~feat_mask) | feat_value;
+ return bd_config->feature_bitmap;
+}
+
+/**
+ * Set the subinterface to run in l2 or l3 mode.
+ * For L3 mode, just the sw_if_index is specified.
+ * For bridged mode, the bd id and bvi flag are also specified.
+ * For xconnect mode, the peer sw_if_index is also specified.
+ * Return 0 if ok, or non-0 if there was an error.
+ */
+
+u32
+set_int_l2_mode (vlib_main_t * vm, vnet_main_t * vnet_main, u32 mode, u32 sw_if_index, u32 bd_index, /* for bridged interface */
+ u32 bvi, /* the bridged interface is the BVI */
+ u32 shg, /* the bridged interface's split horizon group */
+ u32 xc_sw_if_index) /* peer interface for xconnect */
+{
+ l2input_main_t *mp = &l2input_main;
+ l2output_main_t *l2om = &l2output_main;
+ vnet_main_t *vnm = vnet_get_main ();
+ vnet_hw_interface_t *hi;
+ l2_output_config_t *out_config;
+ l2_input_config_t *config;
+ l2_bridge_domain_t *bd_config;
+ l2_flood_member_t member;
+ u64 mac;
+ i32 l2_if_adjust = 0;
+ u32 slot;
+
+ hi = vnet_get_sup_hw_interface (vnet_main, sw_if_index);
+ config = l2input_intf_config (sw_if_index);
+
+ if (config->bridge)
+ {
+ /* Interface is already in bridge mode. Undo the existing config. */
+ bd_config = vec_elt_at_index (mp->bd_configs, config->bd_index);
+
+ /* remove interface from flood vector */
+ bd_remove_member (bd_config, sw_if_index);
+
+ /* undo any BVI-related config */
+ if (bd_config->bvi_sw_if_index == sw_if_index)
+ {
+ bd_config->bvi_sw_if_index = ~0;
+ config->bvi = 0;
+
+ /* delete the l2fib entry for the bvi interface */
+ mac = *((u64 *) hi->hw_address);
+ l2fib_del_entry (mac, config->bd_index);
+
+ /* Make loop output node send packet back to ethernet-input node */
+ slot =
+ vlib_node_add_named_next_with_slot (vm, hi->tx_node_index,
+ "ethernet-input",
+ VNET_SIMULATED_ETHERNET_TX_NEXT_ETHERNET_INPUT);
+ ASSERT (slot == VNET_SIMULATED_ETHERNET_TX_NEXT_ETHERNET_INPUT);
+ }
+ l2_if_adjust--;
+ }
+ else if (config->xconnect)
+ {
+ l2_if_adjust--;
+ }
+
+ /*
+ * Directs the l2 output path to work out the interface
+ * output next-arc itself. Needed when recycling a sw_if_index.
+ */
+ vec_validate_init_empty (l2om->next_nodes.output_node_index_vec,
+ sw_if_index, ~0);
+ l2om->next_nodes.output_node_index_vec[sw_if_index] = ~0;
+
+ /* Initialize the l2-input configuration for the interface */
+ if (mode == MODE_L3)
+ {
+ /* Set L2 config to BD index 0 so that if any packet accidentally
+ * came in on L2 path, it will be dropped in BD 0 */
+ config->xconnect = 0;
+ config->bridge = 0;
+ config->shg = 0;
+ config->bd_index = 0;
+ config->feature_bitmap = L2INPUT_FEAT_DROP;
+
+ /* Make sure any L2-output packet to this interface now in L3 mode is
+ * dropped. This may happen if L2 FIB MAC entry is stale */
+ l2om->next_nodes.output_node_index_vec[sw_if_index] =
+ L2OUTPUT_NEXT_BAD_INTF;
+ }
+ else if (mode == MODE_L2_CLASSIFY)
+ {
+ config->xconnect = 1;
+ config->bridge = 0;
+ config->output_sw_if_index = xc_sw_if_index;
+
+ /* Make sure last-chance drop is configured */
+ config->feature_bitmap |=
+ L2INPUT_FEAT_DROP | L2INPUT_FEAT_INPUT_CLASSIFY;
+
+ /* Make sure bridging features are disabled */
+ config->feature_bitmap &=
+ ~(L2INPUT_FEAT_LEARN | L2INPUT_FEAT_FWD | L2INPUT_FEAT_FLOOD);
+ shg = 0; /* not used in xconnect */
+
+ /* Insure all packets go to ethernet-input */
+ ethernet_set_rx_redirect (vnet_main, hi, 1);
+ }
+ else
+ {
+
+ if (mode == MODE_L2_BRIDGE)
+ {
+ /*
+ * Remove a check that the interface must be an Ethernet.
+ * Specifically so we can bridge to L3 tunnel interfaces.
+ * Here's the check:
+ * if (hi->hw_class_index != ethernet_hw_interface_class.index)
+ *
+ */
+ if (!hi)
+ return MODE_ERROR_ETH; /* non-ethernet */
+
+ config->xconnect = 0;
+ config->bridge = 1;
+ config->bd_index = bd_index;
+
+ /*
+ * Enable forwarding, flooding, learning and ARP termination by default
+ * (note that ARP term is disabled on BD feature bitmap by default)
+ */
+ config->feature_bitmap |= L2INPUT_FEAT_FWD | L2INPUT_FEAT_UU_FLOOD |
+ L2INPUT_FEAT_FLOOD | L2INPUT_FEAT_LEARN | L2INPUT_FEAT_ARP_TERM;
+
+ /* Make sure last-chance drop is configured */
+ config->feature_bitmap |= L2INPUT_FEAT_DROP;
+
+ /* Make sure xconnect is disabled */
+ config->feature_bitmap &= ~L2INPUT_FEAT_XCONNECT;
+
+ /* Set up bridge domain */
+ vec_validate (mp->bd_configs, bd_index);
+ bd_config = vec_elt_at_index (mp->bd_configs, bd_index);
+ bd_validate (bd_config);
+
+ /* TODO: think: add l2fib entry even for non-bvi interface? */
+
+ /* Do BVI interface initializations */
+ if (bvi)
+ {
+ /* ensure BD has no bvi interface (or replace that one with this??) */
+ if (bd_config->bvi_sw_if_index != ~0)
+ {
+ return MODE_ERROR_BVI_DEF; /* bd already has a bvi interface */
+ }
+ bd_config->bvi_sw_if_index = sw_if_index;
+ config->bvi = 1;
+
+ /* create the l2fib entry for the bvi interface */
+ mac = *((u64 *) hi->hw_address);
+ l2fib_add_entry (mac, bd_index, sw_if_index, 1, 0, 1); /* static + bvi */
+
+ /* Disable learning by default. no use since l2fib entry is static. */
+ config->feature_bitmap &= ~L2INPUT_FEAT_LEARN;
+
+ /* Make loop output node send packet to l2-input node */
+ slot =
+ vlib_node_add_named_next_with_slot (vm, hi->tx_node_index,
+ "l2-input",
+ VNET_SIMULATED_ETHERNET_TX_NEXT_ETHERNET_INPUT);
+ ASSERT (slot == VNET_SIMULATED_ETHERNET_TX_NEXT_ETHERNET_INPUT);
+ }
+
+ /* Add interface to bridge-domain flood vector */
+ member.sw_if_index = sw_if_index;
+ member.flags = bvi ? L2_FLOOD_MEMBER_BVI : L2_FLOOD_MEMBER_NORMAL;
+ member.shg = shg;
+ bd_add_member (bd_config, &member);
+
+ }
+ else
+ {
+ config->xconnect = 1;
+ config->bridge = 0;
+ config->output_sw_if_index = xc_sw_if_index;
+
+ /* Make sure last-chance drop is configured */
+ config->feature_bitmap |= L2INPUT_FEAT_DROP;
+
+ /* Make sure bridging features are disabled */
+ config->feature_bitmap &=
+ ~(L2INPUT_FEAT_LEARN | L2INPUT_FEAT_FWD | L2INPUT_FEAT_FLOOD);
+
+ config->feature_bitmap |= L2INPUT_FEAT_XCONNECT;
+ shg = 0; /* not used in xconnect */
+ }
+
+ /* set up split-horizon group */
+ config->shg = shg;
+ out_config = l2output_intf_config (sw_if_index);
+ out_config->shg = shg;
+
+ /*
+ * Test: remove this when non-IP features can be configured.
+ * Enable a non-IP feature to test IP feature masking
+ * config->feature_bitmap |= L2INPUT_FEAT_CTRL_PKT;
+ */
+
+ l2_if_adjust++;
+ }
+
+ /* Adjust count of L2 interfaces */
+ hi->l2_if_count += l2_if_adjust;
+
+ if (hi->hw_class_index == ethernet_hw_interface_class.index)
+ {
+ if ((hi->l2_if_count == 1) && (l2_if_adjust == 1))
+ {
+ /* Just added first L2 interface on this port */
+
+ /* Set promiscuous mode on the l2 interface */
+ ethernet_set_flags (vnet_main, hi->hw_if_index,
+ ETHERNET_INTERFACE_FLAG_ACCEPT_ALL);
+
+ /* ensure all packets go to ethernet-input */
+ ethernet_set_rx_redirect (vnet_main, hi, 1);
+
+ }
+ else if ((hi->l2_if_count == 0) && (l2_if_adjust == -1))
+ {
+ /* Just removed only L2 subinterface on this port */
+
+ /* Disable promiscuous mode on the l2 interface */
+ ethernet_set_flags (vnet_main, hi->hw_if_index, 0);
+
+ /* Allow ip packets to go directly to ip4-input etc */
+ ethernet_set_rx_redirect (vnet_main, hi, 0);
+ }
+ }
+
+ /* Set up the L2/L3 flag in the interface parsing tables */
+ ethernet_sw_interface_set_l2_mode (vnm, sw_if_index, (mode != MODE_L3));
+
+ return 0;
+}
+
+/**
+ * Set subinterface in bridging mode with a bridge-domain ID.
+ * The CLI format is:
+ * set interface l2 bridge <interface> <bd> [bvi] [split-horizon-group]
+ */
+static clib_error_t *
+int_l2_bridge (vlib_main_t * vm,
+ unformat_input_t * input, vlib_cli_command_t * cmd)
+{
+ vnet_main_t *vnm = vnet_get_main ();
+ clib_error_t *error = 0;
+ u32 bd_index, bd_id;
+ u32 sw_if_index;
+ u32 bvi;
+ u32 rc;
+ u32 shg;
+
+ if (!unformat_user (input, unformat_vnet_sw_interface, vnm, &sw_if_index))
+ {
+ error = clib_error_return (0, "unknown interface `%U'",
+ format_unformat_error, input);
+ goto done;
+ }
+
+ if (!unformat (input, "%d", &bd_id))
+ {
+ error = clib_error_return (0, "expected bridge domain ID `%U'",
+ format_unformat_error, input);
+ goto done;
+ }
+
+ bd_index = bd_find_or_add_bd_index (&bd_main, bd_id);
+
+ /* optional bvi */
+ bvi = unformat (input, "bvi");
+
+ /* optional split horizon group */
+ shg = 0;
+ (void) unformat (input, "%d", &shg);
+
+ /* set the interface mode */
+ if ((rc =
+ set_int_l2_mode (vm, vnm, MODE_L2_BRIDGE, sw_if_index, bd_index, bvi,
+ shg, 0)))
+ {
+ if (rc == MODE_ERROR_ETH)
+ {
+ error = clib_error_return (0, "bridged interface must be ethernet",
+ format_unformat_error, input);
+ }
+ else if (rc == MODE_ERROR_BVI_DEF)
+ {
+ error =
+ clib_error_return (0, "bridge-domain already has a bvi interface",
+ format_unformat_error, input);
+ }
+ else
+ {
+ error = clib_error_return (0, "invalid configuration for interface",
+ format_unformat_error, input);
+ }
+ goto done;
+ }
+
+done:
+ return error;
+}
+
+/*?
+ * Use this command put an interface into Layer 2 bridge domain. If a
+ * bridge-domain with the provided bridge-domain-id does not exist, it
+ * will be created. Interfaces in a bridge-domain forward packets to
+ * other interfaces in the same bridge-domain based on destination mac
+ * address. To remove an interface from a the Layer 2 bridge domain,
+ * put the interface in a different mode, for example Layer 3 mode.
+ *
+ * Optionally, an interface can be added to a Layer 2 bridge-domain as
+ * a Bridged Virtual Interface (bvi). Only one interface in a Layer 2
+ * bridge-domain can be a bvi.
+ *
+ * Optionally, a split-horizon group can also be specified. This defaults
+ * to 0 if not specified.
+ *
+ * @cliexpar
+ * Example of how to configure a Layer 2 bridge-domain with three
+ * interfaces (where 200 is the bridge-domain-id):
+ * @cliexcmd{set interface l2 bridge GigabitEthernet0/8/0.200 200}
+ * This interface is added a BVI interface:
+ * @cliexcmd{set interface l2 bridge GigabitEthernet0/9/0.200 200 bvi}
+ * This interface also has a split-horizon group of 1 specified:
+ * @cliexcmd{set interface l2 bridge GigabitEthernet0/a/0.200 200 1}
+ * Example of how to remove an interface from a Layer2 bridge-domain:
+ * @cliexcmd{set interface l3 GigabitEthernet0/a/0.200}
+?*/
+/* *INDENT-OFF* */
+VLIB_CLI_COMMAND (int_l2_bridge_cli, static) = {
+ .path = "set interface l2 bridge",
+ .short_help = "set interface l2 bridge <interface> <bridge-domain-id> [bvi] [shg]",
+ .function = int_l2_bridge,
+};
+/* *INDENT-ON* */
+
+/**
+ * Set subinterface in xconnect mode with another interface.
+ * The CLI format is:
+ * set interface l2 xconnect <interface> <peer interface>
+ */
+static clib_error_t *
+int_l2_xc (vlib_main_t * vm,
+ unformat_input_t * input, vlib_cli_command_t * cmd)
+{
+ vnet_main_t *vnm = vnet_get_main ();
+ clib_error_t *error = 0;
+ u32 sw_if_index;
+ u32 xc_sw_if_index;
+
+ if (!unformat_user (input, unformat_vnet_sw_interface, vnm, &sw_if_index))
+ {
+ error = clib_error_return (0, "unknown interface `%U'",
+ format_unformat_error, input);
+ goto done;
+ }
+
+ if (!unformat_user
+ (input, unformat_vnet_sw_interface, vnm, &xc_sw_if_index))
+ {
+ error = clib_error_return (0, "unknown peer interface `%U'",
+ format_unformat_error, input);
+ goto done;
+ }
+
+ /* set the interface mode */
+ if (set_int_l2_mode
+ (vm, vnm, MODE_L2_XC, sw_if_index, 0, 0, 0, xc_sw_if_index))
+ {
+ error = clib_error_return (0, "invalid configuration for interface",
+ format_unformat_error, input);
+ goto done;
+ }
+
+done:
+ return error;
+}
+
+/*?
+ * Use this command put an interface into Layer 2 cross-connect mode.
+ * Both interfaces must be in this mode for bi-directioal traffic. All
+ * packets received on one interface will be transmitted to the other.
+ * To remove the Layer 2 cross-connect, put the interface in a different
+ * mode, for example Layer 3 mode.
+ *
+ * @cliexpar
+ * Example of how to configure a Layer2 cross-connect between two interfaces:
+ * @cliexcmd{set interface l2 xconnect GigabitEthernet0/8/0.300 GigabitEthernet0/9/0.300}
+ * @cliexcmd{set interface l2 xconnect GigabitEthernet0/9/0.300 GigabitEthernet0/8/0.300}
+ * Example of how to remove a Layer2 cross-connect:
+ * @cliexcmd{set interface l3 GigabitEthernet0/8/0.300}
+ * @cliexcmd{set interface l3 GigabitEthernet0/9/0.300}
+?*/
+/* *INDENT-OFF* */
+VLIB_CLI_COMMAND (int_l2_xc_cli, static) = {
+ .path = "set interface l2 xconnect",
+ .short_help = "set interface l2 xconnect <interface> <peer interface>",
+ .function = int_l2_xc,
+};
+/* *INDENT-ON* */
+
+/**
+ * Set subinterface in L3 mode.
+ * The CLI format is:
+ * set interface l3 <interface>
+ */
+static clib_error_t *
+int_l3 (vlib_main_t * vm, unformat_input_t * input, vlib_cli_command_t * cmd)
+{
+ vnet_main_t *vnm = vnet_get_main ();
+ clib_error_t *error = 0;
+ u32 sw_if_index;
+
+ if (!unformat_user (input, unformat_vnet_sw_interface, vnm, &sw_if_index))
+ {
+ error = clib_error_return (0, "unknown interface `%U'",
+ format_unformat_error, input);
+ goto done;
+ }
+
+ /* set the interface mode */
+ if (set_int_l2_mode (vm, vnm, MODE_L3, sw_if_index, 0, 0, 0, 0))
+ {
+ error = clib_error_return (0, "invalid configuration for interface",
+ format_unformat_error, input);
+ goto done;
+ }
+
+done:
+ return error;
+}
+
+/*?
+ * Modify the packet processing mode of the interface to Layer 3, which
+ * implies packets will be routed. This is the default mode of an interface.
+ * Use this command to remove an interface from a Layer 2 cross-connect or a
+ * Layer 2 bridge.
+ *
+ * @cliexpar
+ * Example of how to set the mode of an interface to Layer 3:
+ * @cliexcmd{set interface l3 GigabitEthernet0/8/0.200}
+?*/
+/* *INDENT-OFF* */
+ VLIB_CLI_COMMAND (int_l3_cli, static) = {
+ .path = "set interface l3",
+ .short_help = "set interface l3 <interface>",
+ .function = int_l3,
+};
+/* *INDENT-ON* */
+
+/**
+ * Show interface mode.
+ * The CLI format is:
+ * show mode [<if-name1> <if-name2> ...]
+ */
+static clib_error_t *
+show_int_mode (vlib_main_t * vm,
+ unformat_input_t * input, vlib_cli_command_t * cmd)
+{
+ vnet_main_t *vnm = vnet_get_main ();
+ clib_error_t *error = 0;
+ char *mode;
+ u8 *args;
+ vnet_interface_main_t *im = &vnm->interface_main;
+ vnet_sw_interface_t *si, *sis = 0;
+ l2input_main_t *mp = &l2input_main;
+ l2_input_config_t *config;
+
+ while (unformat_check_input (input) != UNFORMAT_END_OF_INPUT)
+ {
+ u32 sw_if_index;
+
+ /* See if user wants to show specific interface */
+ if (unformat
+ (input, "%U", unformat_vnet_sw_interface, vnm, &sw_if_index))
+ {
+ si = pool_elt_at_index (im->sw_interfaces, sw_if_index);
+ vec_add1 (sis, si[0]);
+ }
+ else
+ {
+ error = clib_error_return (0, "unknown input `%U'",
+ format_unformat_error, input);
+ goto done;
+ }
+
+ }
+
+ if (vec_len (sis) == 0) /* Get all interfaces */
+ {
+ /* Gather interfaces. */
+ sis = vec_new (vnet_sw_interface_t, pool_elts (im->sw_interfaces));
+ _vec_len (sis) = 0;
+ /* *INDENT-OFF* */
+ pool_foreach (si, im->sw_interfaces, ({ vec_add1 (sis, si[0]); }));
+ /* *INDENT-ON* */
+ }
+
+ vec_foreach (si, sis)
+ {
+ vec_validate (mp->configs, si->sw_if_index);
+ config = vec_elt_at_index (mp->configs, si->sw_if_index);
+ if (config->bridge)
+ {
+ u32 bd_id;
+ mode = "l2 bridge";
+ bd_id = l2input_main.bd_configs[config->bd_index].bd_id;
+
+ args = format (0, "bd_id %d%s%d", bd_id,
+ config->bvi ? " bvi shg " : " shg ", config->shg);
+ }
+ else if (config->xconnect)
+ {
+ mode = "l2 xconnect";
+ args = format (0, "%U",
+ format_vnet_sw_if_index_name,
+ vnm, config->output_sw_if_index);
+ }
+ else
+ {
+ mode = "l3";
+ args = format (0, " ");
+ }
+ vlib_cli_output (vm, "%s %U %v\n",
+ mode,
+ format_vnet_sw_if_index_name,
+ vnm, si->sw_if_index, args);
+ vec_free (args);
+ }
+
+done:
+ vec_free (sis);
+
+ return error;
+}
+
+/*?
+ * Show the packet processing mode (Layer2 xcross-onnect, Layer 2 bridge,
+ * Layer 3 routed) of all interfaces and sub-interfaces, or limit the
+ * output to just the provided list of interfaces and sub-interfaces.
+ * The output shows the mode, the interface, and if the interface is
+ * a member of a bridge, the bridge-domain-id and the split horizen group (shg).
+ *
+ * @cliexpar
+ * Example of displaying the mode of all interfaces:
+ * @cliexstart{show mode}
+ * l3 local0
+ * l3 GigabitEthernet0/8/0
+ * l3 GigabitEthernet0/9/0
+ * l3 GigabitEthernet0/a/0
+ * l2 bridge GigabitEthernet0/8/0.200 bd_id 200 shg 0
+ * l2 bridge GigabitEthernet0/9/0.200 bd_id 200 shg 0
+ * l2 bridge GigabitEthernet0/a/0.200 bd_id 200 shg 0
+ * l2 xconnect GigabitEthernet0/8/0.300 GigabitEthernet0/9/0.300
+ * l2 xconnect GigabitEthernet0/9/0.300 GigabitEthernet0/8/0.300
+ * @cliexend
+ * Example of displaying the mode of a seleted list of interfaces:
+ * @cliexstart{show mode GigabitEthernet0/8/0 GigabitEthernet0/8/0.200}
+ * l3 GigabitEthernet0/8/0
+ * l2 bridge GigabitEthernet0/8/0.200 bd_id 200 shg 0
+ * @cliexend
+?*/
+/* *INDENT-OFF* */
+VLIB_CLI_COMMAND (show_l2_mode, static) = {
+ .path = "show mode",
+ .short_help = "show mode [<if-name1> <if-name2> ...]",
+ .function = show_int_mode,
+};
+/* *INDENT-ON* */
+
+#define foreach_l2_init_function \
+_(feat_bitmap_drop_init) \
+_(l2fib_init) \
+_(l2_input_classify_init) \
+_(l2bd_init) \
+_(l2fwd_init) \
+_(l2_inacl_init) \
+_(l2input_init) \
+_(l2_vtr_init) \
+_(l2_invtr_init) \
+_(l2_efp_filter_init) \
+_(l2learn_init) \
+_(l2flood_init) \
+_(l2_outacl_init) \
+_(l2output_init) \
+_(l2_patch_init) \
+_(l2_xcrw_init)
+
+clib_error_t *
+l2_init (vlib_main_t * vm)
+{
+ clib_error_t *error;
+
+#define _(a) do { \
+ if ((error = vlib_call_init_function (vm, a))) return error; } \
+while (0);
+ foreach_l2_init_function;
+#undef _
+ return 0;
+}
+
+VLIB_INIT_FUNCTION (l2_init);
+
+/*
+ * fd.io coding-style-patch-verification: ON
+ *
+ * Local Variables:
+ * eval: (c-set-style "gnu")
+ * End:
+ */