diff options
Diffstat (limited to 'extras/deprecated/vnet/vxlan-gbp/decap.c')
| -rw-r--r-- | extras/deprecated/vnet/vxlan-gbp/decap.c | 1050 |
1 files changed, 1050 insertions, 0 deletions
diff --git a/extras/deprecated/vnet/vxlan-gbp/decap.c b/extras/deprecated/vnet/vxlan-gbp/decap.c new file mode 100644 index 00000000000..927c778b211 --- /dev/null +++ b/extras/deprecated/vnet/vxlan-gbp/decap.c @@ -0,0 +1,1050 @@ +/* + * decap.c: vxlan gbp tunnel decap packet processing + * + * Copyright (c) 2018 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/vxlan-gbp/vxlan_gbp.h> + +typedef struct +{ + u32 next_index; + u32 tunnel_index; + u32 error; + u32 vni; + u16 sclass; + u8 flags; +} vxlan_gbp_rx_trace_t; + +static u8 * +format_vxlan_gbp_rx_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 *); + vxlan_gbp_rx_trace_t *t = va_arg (*args, vxlan_gbp_rx_trace_t *); + + if (t->tunnel_index == ~0) + return format (s, + "VXLAN_GBP decap error - tunnel for vni %d does not exist", + t->vni); + return format (s, + "VXLAN_GBP decap from vxlan_gbp_tunnel%d vni %d sclass %d" + " flags %U next %d error %d", + t->tunnel_index, t->vni, t->sclass, + format_vxlan_gbp_header_gpflags, t->flags, + t->next_index, t->error); +} + +always_inline u32 +buf_fib_index (vlib_buffer_t * b, u32 is_ip4) +{ + u32 sw_if_index = vnet_buffer (b)->sw_if_index[VLIB_TX]; + if (sw_if_index != (u32) ~ 0) + return sw_if_index; + + u32 *fib_index_by_sw_if_index = is_ip4 ? + ip4_main.fib_index_by_sw_if_index : ip6_main.fib_index_by_sw_if_index; + sw_if_index = vnet_buffer (b)->sw_if_index[VLIB_RX]; + + return vec_elt (fib_index_by_sw_if_index, sw_if_index); +} + +typedef vxlan4_gbp_tunnel_key_t last_tunnel_cache4; + +always_inline vxlan_gbp_tunnel_t * +vxlan4_gbp_find_tunnel (vxlan_gbp_main_t * vxm, last_tunnel_cache4 * cache, + u32 fib_index, ip4_header_t * ip4_0, + vxlan_gbp_header_t * vxlan_gbp0) +{ + /* + * Check unicast first since that's where most of the traffic comes from + * Make sure VXLAN_GBP tunnel exist according to packet SIP, DIP and VNI + */ + vxlan4_gbp_tunnel_key_t key4; + int rv; + + key4.key[1] = (((u64) fib_index << 32) | + (vxlan_gbp0->vni_reserved & + clib_host_to_net_u32 (0xffffff00))); + key4.key[0] = + (((u64) ip4_0->dst_address.as_u32 << 32) | ip4_0->src_address.as_u32); + + if (PREDICT_FALSE (key4.key[0] != cache->key[0] || + key4.key[1] != cache->key[1])) + { + rv = clib_bihash_search_inline_16_8 (&vxm->vxlan4_gbp_tunnel_by_key, + &key4); + if (PREDICT_FALSE (rv == 0)) + { + *cache = key4; + return (pool_elt_at_index (vxm->tunnels, cache->value)); + } + } + else + { + return (pool_elt_at_index (vxm->tunnels, cache->value)); + } + + /* No unicast match - try multicast */ + if (PREDICT_TRUE (!ip4_address_is_multicast (&ip4_0->dst_address))) + return (NULL); + + key4.key[0] = ip4_0->dst_address.as_u32; + /* Make sure mcast VXLAN_GBP tunnel exist by packet DIP and VNI */ + rv = clib_bihash_search_inline_16_8 (&vxm->vxlan4_gbp_tunnel_by_key, &key4); + + if (PREDICT_FALSE (rv != 0)) + return (NULL); + + return (pool_elt_at_index (vxm->tunnels, key4.value)); +} + +typedef vxlan6_gbp_tunnel_key_t last_tunnel_cache6; + +always_inline vxlan_gbp_tunnel_t * +vxlan6_gbp_find_tunnel (vxlan_gbp_main_t * vxm, last_tunnel_cache6 * cache, + u32 fib_index, ip6_header_t * ip6_0, + vxlan_gbp_header_t * vxlan_gbp0) +{ + /* Make sure VXLAN_GBP tunnel exist according to packet SIP and VNI */ + vxlan6_gbp_tunnel_key_t key6 = { + .key = { + [0] = ip6_0->src_address.as_u64[0], + [1] = ip6_0->src_address.as_u64[1], + [2] = ((((u64) fib_index) << 32) | + (vxlan_gbp0->vni_reserved & + clib_host_to_net_u32 (0xffffff00))), + } + }; + int rv; + + if (PREDICT_FALSE + (clib_bihash_key_compare_24_8 (key6.key, cache->key) == 0)) + { + rv = clib_bihash_search_inline_24_8 (&vxm->vxlan6_gbp_tunnel_by_key, + &key6); + if (PREDICT_FALSE (rv != 0)) + return NULL; + + *cache = key6; + } + vxlan_gbp_tunnel_t *t0 = pool_elt_at_index (vxm->tunnels, cache->value); + + /* Validate VXLAN_GBP tunnel SIP against packet DIP */ + if (PREDICT_FALSE + (!ip6_address_is_equal (&ip6_0->dst_address, &t0->src.ip6))) + { + /* try multicast */ + if (PREDICT_TRUE (!ip6_address_is_multicast (&ip6_0->dst_address))) + return 0; + + /* Make sure mcast VXLAN_GBP tunnel exist by packet DIP and VNI */ + key6.key[0] = ip6_0->dst_address.as_u64[0]; + key6.key[1] = ip6_0->dst_address.as_u64[1]; + rv = clib_bihash_search_inline_24_8 (&vxm->vxlan6_gbp_tunnel_by_key, + &key6); + if (PREDICT_FALSE (rv != 0)) + return 0; + + } + + return t0; +} + +always_inline vxlan_gbp_input_next_t +vxlan_gbp_tunnel_get_next (const vxlan_gbp_tunnel_t * t, vlib_buffer_t * b0) +{ + if (VXLAN_GBP_TUNNEL_MODE_L2 == t->mode) + return (VXLAN_GBP_INPUT_NEXT_L2_INPUT); + else + { + ethernet_header_t *e0; + u16 type0; + + e0 = vlib_buffer_get_current (b0); + vlib_buffer_advance (b0, sizeof (*e0)); + type0 = clib_net_to_host_u16 (e0->type); + switch (type0) + { + case ETHERNET_TYPE_IP4: + return (VXLAN_GBP_INPUT_NEXT_IP4_INPUT); + case ETHERNET_TYPE_IP6: + return (VXLAN_GBP_INPUT_NEXT_IP6_INPUT); + } + } + return (VXLAN_GBP_INPUT_NEXT_DROP); +} + +always_inline uword +vxlan_gbp_input (vlib_main_t * vm, + vlib_node_runtime_t * node, + vlib_frame_t * from_frame, u8 is_ip4) +{ + vxlan_gbp_main_t *vxm = &vxlan_gbp_main; + vnet_main_t *vnm = vxm->vnet_main; + vnet_interface_main_t *im = &vnm->interface_main; + vlib_combined_counter_main_t *rx_counter = + im->combined_sw_if_counters + VNET_INTERFACE_COUNTER_RX; + vlib_combined_counter_main_t *drop_counter = + im->combined_sw_if_counters + VNET_INTERFACE_COUNTER_DROP; + last_tunnel_cache4 last4; + last_tunnel_cache6 last6; + u32 pkts_decapsulated = 0; + u32 thread_index = vlib_get_thread_index (); + + if (is_ip4) + clib_memset (&last4, 0xff, sizeof last4); + else + clib_memset (&last6, 0xff, sizeof last6); + + u32 next_index = node->cached_next_index; + + u32 *from = vlib_frame_vector_args (from_frame); + u32 n_left_from = from_frame->n_vectors; + + while (n_left_from > 0) + { + u32 *to_next, n_left_to_next; + vlib_get_next_frame (vm, node, next_index, to_next, n_left_to_next); + + while (n_left_from >= 4 && n_left_to_next >= 2) + { + /* 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); + } + + u32 bi0 = to_next[0] = from[0]; + u32 bi1 = to_next[1] = from[1]; + from += 2; + to_next += 2; + n_left_to_next -= 2; + n_left_from -= 2; + + vlib_buffer_t *b0, *b1; + b0 = vlib_get_buffer (vm, bi0); + b1 = vlib_get_buffer (vm, bi1); + + /* udp leaves current_data pointing at the vxlan_gbp header */ + void *cur0 = vlib_buffer_get_current (b0); + void *cur1 = vlib_buffer_get_current (b1); + vxlan_gbp_header_t *vxlan_gbp0 = cur0; + vxlan_gbp_header_t *vxlan_gbp1 = cur1; + + ip4_header_t *ip4_0, *ip4_1; + ip6_header_t *ip6_0, *ip6_1; + if (is_ip4) + { + ip4_0 = cur0 - sizeof (udp_header_t) - sizeof (ip4_header_t); + ip4_1 = cur1 - sizeof (udp_header_t) - sizeof (ip4_header_t); + } + else + { + ip6_0 = cur0 - sizeof (udp_header_t) - sizeof (ip6_header_t); + ip6_1 = cur1 - sizeof (udp_header_t) - sizeof (ip6_header_t); + } + + u32 fi0 = buf_fib_index (b0, is_ip4); + u32 fi1 = buf_fib_index (b1, is_ip4); + + vxlan_gbp_tunnel_t *t0, *t1; + if (is_ip4) + { + t0 = + vxlan4_gbp_find_tunnel (vxm, &last4, fi0, ip4_0, vxlan_gbp0); + t1 = + vxlan4_gbp_find_tunnel (vxm, &last4, fi1, ip4_1, vxlan_gbp1); + } + else + { + t0 = + vxlan6_gbp_find_tunnel (vxm, &last6, fi0, ip6_0, vxlan_gbp0); + t1 = + vxlan6_gbp_find_tunnel (vxm, &last6, fi1, ip6_1, vxlan_gbp1); + } + + u32 len0 = vlib_buffer_length_in_chain (vm, b0); + u32 len1 = vlib_buffer_length_in_chain (vm, b1); + + vxlan_gbp_input_next_t next0, next1; + u8 error0 = 0, error1 = 0; + u8 flags0 = vxlan_gbp_get_flags (vxlan_gbp0); + u8 flags1 = vxlan_gbp_get_flags (vxlan_gbp1); + /* Required to make the l2 tag push / pop code work on l2 subifs */ + /* pop vxlan_gbp */ + vlib_buffer_advance (b0, sizeof *vxlan_gbp0); + vlib_buffer_advance (b1, sizeof *vxlan_gbp1); + + u8 i_and_g0 = ((flags0 & VXLAN_GBP_FLAGS_GI) == VXLAN_GBP_FLAGS_GI); + u8 i_and_g1 = ((flags1 & VXLAN_GBP_FLAGS_GI) == VXLAN_GBP_FLAGS_GI); + + /* Validate VXLAN_GBP tunnel encap-fib index against packet */ + if (PREDICT_FALSE (t0 == NULL || !i_and_g0)) + { + if (t0 != NULL && !i_and_g0) + { + error0 = VXLAN_GBP_ERROR_BAD_FLAGS; + vlib_increment_combined_counter + (drop_counter, thread_index, t0->sw_if_index, 1, len0); + next0 = VXLAN_GBP_INPUT_NEXT_DROP; + } + else + { + error0 = VXLAN_GBP_ERROR_NO_SUCH_TUNNEL; + next0 = VXLAN_GBP_INPUT_NEXT_PUNT; + if (is_ip4) + b0->punt_reason = + vxm->punt_no_such_tunnel[FIB_PROTOCOL_IP4]; + else + b0->punt_reason = + vxm->punt_no_such_tunnel[FIB_PROTOCOL_IP6]; + } + b0->error = node->errors[error0]; + } + else + { + next0 = vxlan_gbp_tunnel_get_next (t0, b0); + + /* Set packet input sw_if_index to unicast VXLAN tunnel for learning */ + vnet_buffer (b0)->sw_if_index[VLIB_RX] = t0->sw_if_index; + vlib_increment_combined_counter + (rx_counter, thread_index, t0->sw_if_index, 1, len0); + pkts_decapsulated++; + } + + vnet_buffer2 (b0)->gbp.flags = (vxlan_gbp_get_gpflags (vxlan_gbp0) | + VXLAN_GBP_GPFLAGS_R); + vnet_buffer2 (b0)->gbp.sclass = vxlan_gbp_get_sclass (vxlan_gbp0); + + + if (PREDICT_FALSE (t1 == NULL || !i_and_g1)) + { + if (t1 != NULL && !i_and_g1) + { + error1 = VXLAN_GBP_ERROR_BAD_FLAGS; + vlib_increment_combined_counter + (drop_counter, thread_index, t1->sw_if_index, 1, len1); + next1 = VXLAN_GBP_INPUT_NEXT_DROP; + } + else + { + error1 = VXLAN_GBP_ERROR_NO_SUCH_TUNNEL; + next1 = VXLAN_GBP_INPUT_NEXT_PUNT; + if (is_ip4) + b1->punt_reason = + vxm->punt_no_such_tunnel[FIB_PROTOCOL_IP4]; + else + b1->punt_reason = + vxm->punt_no_such_tunnel[FIB_PROTOCOL_IP6]; + } + b1->error = node->errors[error1]; + } + else + { + next1 = vxlan_gbp_tunnel_get_next (t1, b1); + + /* Set packet input sw_if_index to unicast VXLAN_GBP tunnel for learning */ + vnet_buffer (b1)->sw_if_index[VLIB_RX] = t1->sw_if_index; + pkts_decapsulated++; + + vlib_increment_combined_counter + (rx_counter, thread_index, t1->sw_if_index, 1, len1); + } + + vnet_buffer2 (b1)->gbp.flags = (vxlan_gbp_get_gpflags (vxlan_gbp1) | + VXLAN_GBP_GPFLAGS_R); + + vnet_buffer2 (b1)->gbp.sclass = vxlan_gbp_get_sclass (vxlan_gbp1); + + vnet_update_l2_len (b0); + vnet_update_l2_len (b1); + + if (PREDICT_FALSE (b0->flags & VLIB_BUFFER_IS_TRACED)) + { + vxlan_gbp_rx_trace_t *tr = + vlib_add_trace (vm, node, b0, sizeof (*tr)); + tr->next_index = next0; + tr->error = error0; + tr->tunnel_index = t0 == 0 ? ~0 : t0 - vxm->tunnels; + tr->vni = vxlan_gbp_get_vni (vxlan_gbp0); + tr->sclass = vxlan_gbp_get_sclass (vxlan_gbp0); + tr->flags = vxlan_gbp_get_gpflags (vxlan_gbp0); + } + if (PREDICT_FALSE (b1->flags & VLIB_BUFFER_IS_TRACED)) + { + vxlan_gbp_rx_trace_t *tr = + vlib_add_trace (vm, node, b1, sizeof (*tr)); + tr->next_index = next1; + tr->error = error1; + tr->tunnel_index = t1 == 0 ? ~0 : t1 - vxm->tunnels; + tr->vni = vxlan_gbp_get_vni (vxlan_gbp1); + tr->sclass = vxlan_gbp_get_sclass (vxlan_gbp1); + tr->flags = vxlan_gbp_get_gpflags (vxlan_gbp1); + } + + vlib_validate_buffer_enqueue_x2 (vm, node, next_index, + to_next, n_left_to_next, + bi0, bi1, next0, next1); + } + + while (n_left_from > 0 && n_left_to_next > 0) + { + u32 bi0 = to_next[0] = from[0]; + from += 1; + to_next += 1; + n_left_from -= 1; + n_left_to_next -= 1; + + vlib_buffer_t *b0 = vlib_get_buffer (vm, bi0); + + /* udp leaves current_data pointing at the vxlan_gbp header */ + void *cur0 = vlib_buffer_get_current (b0); + vxlan_gbp_header_t *vxlan_gbp0 = cur0; + ip4_header_t *ip4_0; + ip6_header_t *ip6_0; + if (is_ip4) + ip4_0 = cur0 - sizeof (udp_header_t) - sizeof (ip4_header_t); + else + ip6_0 = cur0 - sizeof (udp_header_t) - sizeof (ip6_header_t); + + u32 fi0 = buf_fib_index (b0, is_ip4); + + vxlan_gbp_tunnel_t *t0; + if (is_ip4) + t0 = vxlan4_gbp_find_tunnel (vxm, &last4, fi0, ip4_0, vxlan_gbp0); + else + t0 = vxlan6_gbp_find_tunnel (vxm, &last6, fi0, ip6_0, vxlan_gbp0); + + uword len0 = vlib_buffer_length_in_chain (vm, b0); + + vxlan_gbp_input_next_t next0; + u8 error0 = 0; + u8 flags0 = vxlan_gbp_get_flags (vxlan_gbp0); + + /* pop (ip, udp, vxlan_gbp) */ + vlib_buffer_advance (b0, sizeof (*vxlan_gbp0)); + + u8 i_and_g0 = ((flags0 & VXLAN_GBP_FLAGS_GI) == VXLAN_GBP_FLAGS_GI); + + /* Validate VXLAN_GBP tunnel encap-fib index against packet */ + if (PREDICT_FALSE (t0 == NULL || !i_and_g0)) + { + if (t0 != NULL && !i_and_g0) + { + error0 = VXLAN_GBP_ERROR_BAD_FLAGS; + vlib_increment_combined_counter + (drop_counter, thread_index, t0->sw_if_index, 1, len0); + next0 = VXLAN_GBP_INPUT_NEXT_DROP; + } + else + { + error0 = VXLAN_GBP_ERROR_NO_SUCH_TUNNEL; + next0 = VXLAN_GBP_INPUT_NEXT_PUNT; + if (is_ip4) + b0->punt_reason = + vxm->punt_no_such_tunnel[FIB_PROTOCOL_IP4]; + else + b0->punt_reason = + vxm->punt_no_such_tunnel[FIB_PROTOCOL_IP6]; + } + b0->error = node->errors[error0]; + } + else + { + next0 = vxlan_gbp_tunnel_get_next (t0, b0); + /* Set packet input sw_if_index to unicast VXLAN_GBP tunnel for learning */ + vnet_buffer (b0)->sw_if_index[VLIB_RX] = t0->sw_if_index; + pkts_decapsulated++; + + vlib_increment_combined_counter + (rx_counter, thread_index, t0->sw_if_index, 1, len0); + } + vnet_buffer2 (b0)->gbp.flags = (vxlan_gbp_get_gpflags (vxlan_gbp0) | + VXLAN_GBP_GPFLAGS_R); + + vnet_buffer2 (b0)->gbp.sclass = vxlan_gbp_get_sclass (vxlan_gbp0); + + /* Required to make the l2 tag push / pop code work on l2 subifs */ + vnet_update_l2_len (b0); + + if (PREDICT_FALSE (b0->flags & VLIB_BUFFER_IS_TRACED)) + { + vxlan_gbp_rx_trace_t *tr + = vlib_add_trace (vm, node, b0, sizeof (*tr)); + tr->next_index = next0; + tr->error = error0; + tr->tunnel_index = t0 == 0 ? ~0 : t0 - vxm->tunnels; + tr->vni = vxlan_gbp_get_vni (vxlan_gbp0); + tr->sclass = vxlan_gbp_get_sclass (vxlan_gbp0); + tr->flags = vxlan_gbp_get_gpflags (vxlan_gbp0); + } + 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); + } + /* Do we still need this now that tunnel tx stats is kept? */ + u32 node_idx = + is_ip4 ? vxlan4_gbp_input_node.index : vxlan6_gbp_input_node.index; + vlib_node_increment_counter (vm, node_idx, VXLAN_GBP_ERROR_DECAPSULATED, + pkts_decapsulated); + + return from_frame->n_vectors; +} + +VLIB_NODE_FN (vxlan4_gbp_input_node) (vlib_main_t * vm, + vlib_node_runtime_t * node, + vlib_frame_t * from_frame) +{ + return vxlan_gbp_input (vm, node, from_frame, /* is_ip4 */ 1); +} + +VLIB_NODE_FN (vxlan6_gbp_input_node) (vlib_main_t * vm, + vlib_node_runtime_t * node, + vlib_frame_t * from_frame) +{ + return vxlan_gbp_input (vm, node, from_frame, /* is_ip4 */ 0); +} + +static char *vxlan_gbp_error_strings[] = { +#define vxlan_gbp_error(n,s) s, +#include <vnet/vxlan-gbp/vxlan_gbp_error.def> +#undef vxlan_gbp_error +#undef _ +}; + +/* *INDENT-OFF* */ +VLIB_REGISTER_NODE (vxlan4_gbp_input_node) = +{ + .name = "vxlan4-gbp-input", + .vector_size = sizeof (u32), + .n_errors = VXLAN_GBP_N_ERROR, + .error_strings = vxlan_gbp_error_strings, + .n_next_nodes = VXLAN_GBP_INPUT_N_NEXT, + .format_trace = format_vxlan_gbp_rx_trace, + .next_nodes = { +#define _(s,n) [VXLAN_GBP_INPUT_NEXT_##s] = n, + foreach_vxlan_gbp_input_next +#undef _ + }, +}; + +VLIB_REGISTER_NODE (vxlan6_gbp_input_node) = +{ + .name = "vxlan6-gbp-input", + .vector_size = sizeof (u32), + .n_errors = VXLAN_GBP_N_ERROR, + .error_strings = vxlan_gbp_error_strings, + .n_next_nodes = VXLAN_GBP_INPUT_N_NEXT, + .next_nodes = { +#define _(s,n) [VXLAN_GBP_INPUT_NEXT_##s] = n, + foreach_vxlan_gbp_input_next +#undef _ + }, + .format_trace = format_vxlan_gbp_rx_trace, +}; +/* *INDENT-ON* */ + +typedef enum +{ + IP_VXLAN_GBP_BYPASS_NEXT_DROP, + IP_VXLAN_GBP_BYPASS_NEXT_VXLAN_GBP, + IP_VXLAN_GBP_BYPASS_N_NEXT, +} ip_vxlan_gbp_bypass_next_t; + +always_inline uword +ip_vxlan_gbp_bypass_inline (vlib_main_t * vm, + vlib_node_runtime_t * node, + vlib_frame_t * frame, u32 is_ip4) +{ + vxlan_gbp_main_t *vxm = &vxlan_gbp_main; + u32 *from, *to_next, n_left_from, n_left_to_next, next_index; + vlib_node_runtime_t *error_node = + vlib_node_get_runtime (vm, ip4_input_node.index); + ip4_address_t addr4; /* last IPv4 address matching a local VTEP address */ + ip6_address_t addr6; /* last IPv6 address matching a local VTEP address */ + + from = vlib_frame_vector_args (frame); + n_left_from = frame->n_vectors; + next_index = node->cached_next_index; + + if (node->flags & VLIB_NODE_FLAG_TRACE) + ip4_forward_next_trace (vm, node, frame, VLIB_TX); + + if (is_ip4) + addr4.data_u32 = ~0; + else + ip6_address_set_zero (&addr6); + + while (n_left_from > 0) + { + vlib_get_next_frame (vm, node, next_index, to_next, n_left_to_next); + + while (n_left_from >= 4 && n_left_to_next >= 2) + { + vlib_buffer_t *b0, *b1; + ip4_header_t *ip40, *ip41; + ip6_header_t *ip60, *ip61; + udp_header_t *udp0, *udp1; + u32 bi0, ip_len0, udp_len0, flags0, next0; + u32 bi1, ip_len1, udp_len1, flags1, next1; + i32 len_diff0, len_diff1; + u8 error0, good_udp0, proto0; + u8 error1, good_udp1, proto1; + + /* 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 = to_next[0] = from[0]; + bi1 = to_next[1] = from[1]; + from += 2; + n_left_from -= 2; + to_next += 2; + n_left_to_next -= 2; + + b0 = vlib_get_buffer (vm, bi0); + b1 = vlib_get_buffer (vm, bi1); + if (is_ip4) + { + ip40 = vlib_buffer_get_current (b0); + ip41 = vlib_buffer_get_current (b1); + } + else + { + ip60 = vlib_buffer_get_current (b0); + ip61 = vlib_buffer_get_current (b1); + } + + /* Setup packet for next IP feature */ + vnet_feature_next (&next0, b0); + vnet_feature_next (&next1, b1); + + if (is_ip4) + { + /* Treat IP frag packets as "experimental" protocol for now + until support of IP frag reassembly is implemented */ + proto0 = ip4_is_fragment (ip40) ? 0xfe : ip40->protocol; + proto1 = ip4_is_fragment (ip41) ? 0xfe : ip41->protocol; + } + else + { + proto0 = ip60->protocol; + proto1 = ip61->protocol; + } + + /* Process packet 0 */ + if (proto0 != IP_PROTOCOL_UDP) + goto exit0; /* not UDP packet */ + + if (is_ip4) + udp0 = ip4_next_header (ip40); + else + udp0 = ip6_next_header (ip60); + + if (udp0->dst_port != clib_host_to_net_u16 (UDP_DST_PORT_vxlan_gbp)) + goto exit0; /* not VXLAN_GBP packet */ + + /* Validate DIP against VTEPs */ + if (is_ip4) + { + if (addr4.as_u32 != ip40->dst_address.as_u32) + { + if (!hash_get (vxm->vtep4, ip40->dst_address.as_u32)) + goto exit0; /* no local VTEP for VXLAN_GBP packet */ + addr4 = ip40->dst_address; + } + } + else + { + if (!ip6_address_is_equal (&addr6, &ip60->dst_address)) + { + if (!hash_get_mem (vxm->vtep6, &ip60->dst_address)) + goto exit0; /* no local VTEP for VXLAN_GBP packet */ + addr6 = ip60->dst_address; + } + } + + flags0 = b0->flags; + good_udp0 = (flags0 & VNET_BUFFER_F_L4_CHECKSUM_CORRECT) != 0; + + /* Don't verify UDP checksum for packets with explicit zero checksum. */ + good_udp0 |= udp0->checksum == 0; + + /* Verify UDP length */ + if (is_ip4) + ip_len0 = clib_net_to_host_u16 (ip40->length); + else + ip_len0 = clib_net_to_host_u16 (ip60->payload_length); + udp_len0 = clib_net_to_host_u16 (udp0->length); + len_diff0 = ip_len0 - udp_len0; + + /* Verify UDP checksum */ + if (PREDICT_FALSE (!good_udp0)) + { + if ((flags0 & VNET_BUFFER_F_L4_CHECKSUM_COMPUTED) == 0) + { + if (is_ip4) + flags0 = ip4_tcp_udp_validate_checksum (vm, b0); + else + flags0 = ip6_tcp_udp_icmp_validate_checksum (vm, b0); + good_udp0 = + (flags0 & VNET_BUFFER_F_L4_CHECKSUM_CORRECT) != 0; + } + } + + if (is_ip4) + { + error0 = good_udp0 ? 0 : IP4_ERROR_UDP_CHECKSUM; + error0 = (len_diff0 >= 0) ? error0 : IP4_ERROR_UDP_LENGTH; + } + else + { + error0 = good_udp0 ? 0 : IP6_ERROR_UDP_CHECKSUM; + error0 = (len_diff0 >= 0) ? error0 : IP6_ERROR_UDP_LENGTH; + } + + next0 = error0 ? + IP_VXLAN_GBP_BYPASS_NEXT_DROP : + IP_VXLAN_GBP_BYPASS_NEXT_VXLAN_GBP; + b0->error = error0 ? error_node->errors[error0] : 0; + + /* vxlan-gbp-input node expect current at VXLAN_GBP header */ + if (is_ip4) + vlib_buffer_advance (b0, + sizeof (ip4_header_t) + + sizeof (udp_header_t)); + else + vlib_buffer_advance (b0, + sizeof (ip6_header_t) + + sizeof (udp_header_t)); + + exit0: + /* Process packet 1 */ + if (proto1 != IP_PROTOCOL_UDP) + goto exit1; /* not UDP packet */ + + if (is_ip4) + udp1 = ip4_next_header (ip41); + else + udp1 = ip6_next_header (ip61); + + if (udp1->dst_port != clib_host_to_net_u16 (UDP_DST_PORT_vxlan_gbp)) + goto exit1; /* not VXLAN_GBP packet */ + + /* Validate DIP against VTEPs */ + if (is_ip4) + { + if (addr4.as_u32 != ip41->dst_address.as_u32) + { + if (!hash_get (vxm->vtep4, ip41->dst_address.as_u32)) + goto exit1; /* no local VTEP for VXLAN_GBP packet */ + addr4 = ip41->dst_address; + } + } + else + { + if (!ip6_address_is_equal (&addr6, &ip61->dst_address)) + { + if (!hash_get_mem (vxm->vtep6, &ip61->dst_address)) + goto exit1; /* no local VTEP for VXLAN_GBP packet */ + addr6 = ip61->dst_address; + } + } + + flags1 = b1->flags; + good_udp1 = (flags1 & VNET_BUFFER_F_L4_CHECKSUM_CORRECT) != 0; + + /* Don't verify UDP checksum for packets with explicit zero checksum. */ + good_udp1 |= udp1->checksum == 0; + + /* Verify UDP length */ + if (is_ip4) + ip_len1 = clib_net_to_host_u16 (ip41->length); + else + ip_len1 = clib_net_to_host_u16 (ip61->payload_length); + udp_len1 = clib_net_to_host_u16 (udp1->length); + len_diff1 = ip_len1 - udp_len1; + + /* Verify UDP checksum */ + if (PREDICT_FALSE (!good_udp1)) + { + if ((flags1 & VNET_BUFFER_F_L4_CHECKSUM_COMPUTED) == 0) + { + if (is_ip4) + flags1 = ip4_tcp_udp_validate_checksum (vm, b1); + else + flags1 = ip6_tcp_udp_icmp_validate_checksum (vm, b1); + good_udp1 = + (flags1 & VNET_BUFFER_F_L4_CHECKSUM_CORRECT) != 0; + } + } + + if (is_ip4) + { + error1 = good_udp1 ? 0 : IP4_ERROR_UDP_CHECKSUM; + error1 = (len_diff1 >= 0) ? error1 : IP4_ERROR_UDP_LENGTH; + } + else + { + error1 = good_udp1 ? 0 : IP6_ERROR_UDP_CHECKSUM; + error1 = (len_diff1 >= 0) ? error1 : IP6_ERROR_UDP_LENGTH; + } + + next1 = error1 ? + IP_VXLAN_GBP_BYPASS_NEXT_DROP : + IP_VXLAN_GBP_BYPASS_NEXT_VXLAN_GBP; + b1->error = error1 ? error_node->errors[error1] : 0; + + /* vxlan_gbp-input node expect current at VXLAN_GBP header */ + if (is_ip4) + vlib_buffer_advance (b1, + sizeof (ip4_header_t) + + sizeof (udp_header_t)); + else + vlib_buffer_advance (b1, + sizeof (ip6_header_t) + + sizeof (udp_header_t)); + + exit1: + vlib_validate_buffer_enqueue_x2 (vm, node, next_index, + to_next, n_left_to_next, + bi0, bi1, next0, next1); + } + + while (n_left_from > 0 && n_left_to_next > 0) + { + vlib_buffer_t *b0; + ip4_header_t *ip40; + ip6_header_t *ip60; + udp_header_t *udp0; + u32 bi0, ip_len0, udp_len0, flags0, next0; + i32 len_diff0; + u8 error0, good_udp0, proto0; + + bi0 = to_next[0] = from[0]; + from += 1; + n_left_from -= 1; + to_next += 1; + n_left_to_next -= 1; + + b0 = vlib_get_buffer (vm, bi0); + if (is_ip4) + ip40 = vlib_buffer_get_current (b0); + else + ip60 = vlib_buffer_get_current (b0); + + /* Setup packet for next IP feature */ + vnet_feature_next (&next0, b0); + + if (is_ip4) + /* Treat IP4 frag packets as "experimental" protocol for now + until support of IP frag reassembly is implemented */ + proto0 = ip4_is_fragment (ip40) ? 0xfe : ip40->protocol; + else + proto0 = ip60->protocol; + + if (proto0 != IP_PROTOCOL_UDP) + goto exit; /* not UDP packet */ + + if (is_ip4) + udp0 = ip4_next_header (ip40); + else + udp0 = ip6_next_header (ip60); + + if (udp0->dst_port != clib_host_to_net_u16 (UDP_DST_PORT_vxlan_gbp)) + goto exit; /* not VXLAN_GBP packet */ + + /* Validate DIP against VTEPs */ + if (is_ip4) + { + if (addr4.as_u32 != ip40->dst_address.as_u32) + { + if (!hash_get (vxm->vtep4, ip40->dst_address.as_u32)) + goto exit; /* no local VTEP for VXLAN_GBP packet */ + addr4 = ip40->dst_address; + } + } + else + { + if (!ip6_address_is_equal (&addr6, &ip60->dst_address)) + { + if (!hash_get_mem (vxm->vtep6, &ip60->dst_address)) + goto exit; /* no local VTEP for VXLAN_GBP packet */ + addr6 = ip60->dst_address; + } + } + + flags0 = b0->flags; + good_udp0 = (flags0 & VNET_BUFFER_F_L4_CHECKSUM_CORRECT) != 0; + + /* Don't verify UDP checksum for packets with explicit zero checksum. */ + good_udp0 |= udp0->checksum == 0; + + /* Verify UDP length */ + if (is_ip4) + ip_len0 = clib_net_to_host_u16 (ip40->length); + else + ip_len0 = clib_net_to_host_u16 (ip60->payload_length); + udp_len0 = clib_net_to_host_u16 (udp0->length); + len_diff0 = ip_len0 - udp_len0; + + /* Verify UDP checksum */ + if (PREDICT_FALSE (!good_udp0)) + { + if ((flags0 & VNET_BUFFER_F_L4_CHECKSUM_COMPUTED) == 0) + { + if (is_ip4) + flags0 = ip4_tcp_udp_validate_checksum (vm, b0); + else + flags0 = ip6_tcp_udp_icmp_validate_checksum (vm, b0); + good_udp0 = + (flags0 & VNET_BUFFER_F_L4_CHECKSUM_CORRECT) != 0; + } + } + + if (is_ip4) + { + error0 = good_udp0 ? 0 : IP4_ERROR_UDP_CHECKSUM; + error0 = (len_diff0 >= 0) ? error0 : IP4_ERROR_UDP_LENGTH; + } + else + { + error0 = good_udp0 ? 0 : IP6_ERROR_UDP_CHECKSUM; + error0 = (len_diff0 >= 0) ? error0 : IP6_ERROR_UDP_LENGTH; + } + + next0 = error0 ? + IP_VXLAN_GBP_BYPASS_NEXT_DROP : + IP_VXLAN_GBP_BYPASS_NEXT_VXLAN_GBP; + b0->error = error0 ? error_node->errors[error0] : 0; + + /* vxlan_gbp-input node expect current at VXLAN_GBP header */ + if (is_ip4) + vlib_buffer_advance (b0, + sizeof (ip4_header_t) + + sizeof (udp_header_t)); + else + vlib_buffer_advance (b0, + sizeof (ip6_header_t) + + sizeof (udp_header_t)); + + exit: + 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; +} + +VLIB_NODE_FN (ip4_vxlan_gbp_bypass_node) (vlib_main_t * vm, + vlib_node_runtime_t * node, + vlib_frame_t * frame) +{ + return ip_vxlan_gbp_bypass_inline (vm, node, frame, /* is_ip4 */ 1); +} + +/* *INDENT-OFF* */ +VLIB_REGISTER_NODE (ip4_vxlan_gbp_bypass_node) = +{ + .name = "ip4-vxlan-gbp-bypass", + .vector_size = sizeof (u32), + .n_next_nodes = IP_VXLAN_GBP_BYPASS_N_NEXT, + .next_nodes = { + [IP_VXLAN_GBP_BYPASS_NEXT_DROP] = "error-drop", + [IP_VXLAN_GBP_BYPASS_NEXT_VXLAN_GBP] = "vxlan4-gbp-input", + }, + .format_buffer = format_ip4_header, + .format_trace = format_ip4_forward_next_trace, +}; +/* *INDENT-ON* */ + +#ifndef CLIB_MARCH_VARIANT +/* Dummy init function to get us linked in. */ +clib_error_t * +ip4_vxlan_gbp_bypass_init (vlib_main_t * vm) +{ + return 0; +} + +VLIB_INIT_FUNCTION (ip4_vxlan_gbp_bypass_init); +#endif /* CLIB_MARCH_VARIANT */ + +VLIB_NODE_FN (ip6_vxlan_gbp_bypass_node) (vlib_main_t * vm, + vlib_node_runtime_t * node, + vlib_frame_t * frame) +{ + return ip_vxlan_gbp_bypass_inline (vm, node, frame, /* is_ip4 */ 0); +} + +/* *INDENT-OFF* */ +VLIB_REGISTER_NODE (ip6_vxlan_gbp_bypass_node) = +{ + .name = "ip6-vxlan-gbp-bypass", + .vector_size = sizeof (u32), + .n_next_nodes = IP_VXLAN_GBP_BYPASS_N_NEXT, + .next_nodes = { + [IP_VXLAN_GBP_BYPASS_NEXT_DROP] = "error-drop", + [IP_VXLAN_GBP_BYPASS_NEXT_VXLAN_GBP] = "vxlan6-gbp-input", + }, + .format_buffer = format_ip6_header, + .format_trace = format_ip6_forward_next_trace, +}; +/* *INDENT-ON* */ + +#ifndef CLIB_MARCH_VARIANT +/* Dummy init function to get us linked in. */ +clib_error_t * +ip6_vxlan_gbp_bypass_init (vlib_main_t * vm) +{ + return 0; +} + +VLIB_INIT_FUNCTION (ip6_vxlan_gbp_bypass_init); +#endif /* CLIB_MARCH_VARIANT */ + +/* + * fd.io coding-style-patch-verification: ON + * + * Local Variables: + * eval: (c-set-style "gnu") + * End: + */ |