diff options
Diffstat (limited to 'src/plugins')
-rw-r--r-- | src/plugins/Makefile.am | 60 | ||||
-rw-r--r-- | src/plugins/flowperpkt.am | 38 | ||||
-rw-r--r-- | src/plugins/flowperpkt/flowperpkt.api | 42 | ||||
-rw-r--r-- | src/plugins/flowperpkt/flowperpkt.c | 671 | ||||
-rw-r--r-- | src/plugins/flowperpkt/flowperpkt.h | 90 | ||||
-rw-r--r-- | src/plugins/flowperpkt/flowperpkt_all_api_h.h | 18 | ||||
-rw-r--r-- | src/plugins/flowperpkt/flowperpkt_msg_enum.h | 31 | ||||
-rw-r--r-- | src/plugins/flowperpkt/flowperpkt_plugin_doc.md | 13 | ||||
-rw-r--r-- | src/plugins/flowperpkt/flowperpkt_test.c | 234 | ||||
-rw-r--r-- | src/plugins/flowperpkt/l2_node.c | 561 | ||||
-rw-r--r-- | src/plugins/flowperpkt/node.c | 574 | ||||
-rw-r--r-- | src/plugins/ila.am | 20 | ||||
-rw-r--r-- | src/plugins/ila/ila.c | 1070 | ||||
-rw-r--r-- | src/plugins/ila/ila.h | 116 | ||||
-rw-r--r-- | src/plugins/sixrd.am | 26 | ||||
-rw-r--r-- | src/plugins/sixrd/ip4_sixrd.c | 127 | ||||
-rw-r--r-- | src/plugins/sixrd/ip6_sixrd.c | 129 | ||||
-rw-r--r-- | src/plugins/sixrd/sixrd.c | 369 | ||||
-rw-r--r-- | src/plugins/sixrd/sixrd.h | 141 | ||||
-rw-r--r-- | src/plugins/sixrd/sixrd_dpo.c | 132 | ||||
-rw-r--r-- | src/plugins/sixrd/sixrd_dpo.h | 61 |
21 files changed, 4523 insertions, 0 deletions
diff --git a/src/plugins/Makefile.am b/src/plugins/Makefile.am new file mode 100644 index 00000000000..ffc4b3abddc --- /dev/null +++ b/src/plugins/Makefile.am @@ -0,0 +1,60 @@ + +# Copyright (c) <current-year> <your-organization> +# 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. + +AUTOMAKE_OPTIONS = foreign subdir-objects + +AM_CFLAGS = -Wall -I${top_srcdir} -I${top_builddir} +AM_LDFLAGS = -module -shared -avoid-version +SUFFIXES = .api.h .api .api.json +API_FILES = +BUILT_SOURCES = +vppplugins_LTLIBRARIES = +vppapitestplugins_LTLIBRARIES = +noinst_HEADERS = + +vppapitestpluginsdir = ${libdir}/vpp_api_test_plugins +vpppluginsdir = ${libdir}/vpp_plugins + +if ENABLE_FLOWPERPKT_PLUGIN +include flowperpkt.am +endif + +if ENABLE_ILA_PLUGIN +include ila.am +endif + +if ENABLE_SIXRD_PLUGIN +include sixrd.am +endif + +include ../suffix-rules.mk + +# Remove *.la files +install-data-hook: + @-(cd $(vpppluginsdir) && $(RM) $(vppplugins_LTLIBRARIES)) + @-(cd $(vppapitestpluginsdir) && $(RM) $(vppapitestplugins_LTLIBRARIES)) + +############################################################################### +# API +############################################################################### + +apidir = $(prefix)/share/vpp/api/plugins + +api_DATA = \ + $(patsubst %.api,%.api.json,$(API_FILES)) + +BUILT_SOURCES += \ + $(patsubst %.api,%.api.json,$(API_FILES)) \ + $(patsubst %.api,%.api.h,$(API_FILES)) + diff --git a/src/plugins/flowperpkt.am b/src/plugins/flowperpkt.am new file mode 100644 index 00000000000..a400603a71f --- /dev/null +++ b/src/plugins/flowperpkt.am @@ -0,0 +1,38 @@ + +# Copyright (c) <current-year> <your-organization> +# 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. + +vppplugins_LTLIBRARIES += flowperpkt_plugin.la +vppapitestplugins_LTLIBRARIES += flowperpkt_test_plugin.la + +flowperpkt_plugin_la_SOURCES = flowperpkt/flowperpkt.c \ + flowperpkt/l2_node.c \ + flowperpkt/node.c \ + flowperpkt/flowperpkt_plugin.api.h + +BUILT_SOURCES += \ + flowperpkt/flowperpkt.api.h \ + flowperpkt/flowperpkt.api.json + +noinst_HEADERS += \ + flowperpkt/flowperpkt_all_api_h.h \ + flowperpkt/flowperpkt_msg_enum.h \ + flowperpkt/flowperpkt.api.h + +flowperpkt_test_plugin_la_SOURCES = \ + flowperpkt/flowperpkt_test.c \ + flowperpkt/flowperpkt_plugin.api.h + +API_FILES += flowperpkt/flowperpkt.api + +# vi:syntax=automake diff --git a/src/plugins/flowperpkt/flowperpkt.api b/src/plugins/flowperpkt/flowperpkt.api new file mode 100644 index 00000000000..fa878f21ed3 --- /dev/null +++ b/src/plugins/flowperpkt/flowperpkt.api @@ -0,0 +1,42 @@ +/* Define a simple enable-disable binary API to control the feature */ + +/** \file + This file defines the vpp control-plane API messages + used to control the flowperpkt plugin +*/ + +/** \brief Enable / disable per-packet IPFIX recording on an interface + @param client_index - opaque cookie to identify the sender + @param context - sender context, to match reply w/ request + @param is_add - add address if non-zero, else delete + @param is_ipv6 - if non-zero the address is ipv6, else ipv4 + @param sw_if_index - index of the interface +*/ +manual_print define flowperpkt_tx_interface_add_del +{ + /* Client identifier, set from api_main.my_client_index */ + u32 client_index; + + /* Arbitrary context, so client can match reply to request */ + u32 context; + + /* Enable / disable the feature */ + u8 is_add; + u8 which; /* 0 = ipv4, 1 = l2, 2 = ipv6 (not yet implemented) */ + + /* Interface handle */ + u32 sw_if_index; +}; + +/** \brief Reply to enable/disable per-packet IPFIX recording messages + @param context - returned sender context, to match reply w/ request + @param retval - return code +*/ +define flowperpkt_tx_interface_add_del_reply +{ + /* From the request */ + u32 context; + + /* Return value, zero means all OK */ + i32 retval; +}; diff --git a/src/plugins/flowperpkt/flowperpkt.c b/src/plugins/flowperpkt/flowperpkt.c new file mode 100644 index 00000000000..fb71d5b0ffc --- /dev/null +++ b/src/plugins/flowperpkt/flowperpkt.c @@ -0,0 +1,671 @@ +/* + * flowperpkt.c - per-packet data capture flow report plugin + * + * Copyright (c) <current-year> <your-organization> + * 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. + */ + +/** + * @file + * @brief Per-packet IPFIX flow record generator plugin + * + * This file implements vpp plugin registration mechanics, + * debug CLI, and binary API handling. + */ + +#include <vnet/vnet.h> +#include <vnet/plugin/plugin.h> +#include <flowperpkt/flowperpkt.h> + +#include <vlibapi/api.h> +#include <vlibmemory/api.h> +#include <vlibsocket/api.h> + +/* define message IDs */ +#include <flowperpkt/flowperpkt_msg_enum.h> + +/* define message structures */ +#define vl_typedefs +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_typedefs + +/* define generated endian-swappers */ +#define vl_endianfun +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_endianfun + +/* instantiate all the print functions we know about */ +#define vl_print(handle, ...) vlib_cli_output (handle, __VA_ARGS__) +#define vl_printfun +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_printfun + +flowperpkt_main_t flowperpkt_main; + +/* Get the API version number */ +#define vl_api_version(n,v) static u32 api_version=(v); +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_api_version + +/* Define the per-interface configurable features */ +/* *INDENT-OFF* */ +VNET_FEATURE_INIT (flow_perpacket_ipv4, static) = +{ + .arc_name = "ip4-output", + .node_name = "flowperpkt-ipv4", + .runs_before = VNET_FEATURES ("interface-output"), +}; + +VNET_FEATURE_INIT (flow_perpacket_l2, static) = +{ + .arc_name = "interface-output", + .node_name = "flowperpkt-l2", + .runs_before = VNET_FEATURES ("interface-tx"), +}; +/* *INDENT-ON* */ + +/* + * A handy macro to set up a message reply. + * Assumes that the following variables are available: + * mp - pointer to request message + * rmp - pointer to reply message type + * rv - return value + */ +#define REPLY_MACRO(t) \ +do { \ + unix_shared_memory_queue_t * q = \ + vl_api_client_index_to_input_queue (mp->client_index); \ + if (!q) \ + return; \ + \ + rmp = vl_msg_api_alloc (sizeof (*rmp)); \ + rmp->_vl_msg_id = ntohs((t)+fm->msg_id_base); \ + rmp->context = mp->context; \ + rmp->retval = ntohl(rv); \ + \ + vl_msg_api_send_shmem (q, (u8 *)&rmp); \ +} while(0); + +/* Macro to finish up custom dump fns */ +#define FINISH \ + vec_add1 (s, 0); \ + vl_print (handle, (char *)s); \ + vec_free (s); \ + return handle; + +#define VALIDATE_SW_IF_INDEX(mp) \ + do { u32 __sw_if_index = ntohl(mp->sw_if_index); \ + vnet_main_t *__vnm = vnet_get_main(); \ + if (pool_is_free_index(__vnm->interface_main.sw_interfaces, \ + __sw_if_index)) { \ + rv = VNET_API_ERROR_INVALID_SW_IF_INDEX; \ + goto bad_sw_if_index; \ + } \ +} while(0); + +#define BAD_SW_IF_INDEX_LABEL \ +do { \ +bad_sw_if_index: \ + ; \ +} while (0); + +/** + * @brief Create an IPFIX template packet rewrite string + * @param frm flow_report_main_t * + * @param fr flow_report_t * + * @param collector_address ip4_address_t * the IPFIX collector address + * @param src_address ip4_address_t * the source address we should use + * @param collector_port u16 the collector port we should use, host byte order + * @returns u8 * vector containing the indicated IPFIX template packet + */ +static inline u8 * +flowperpkt_template_rewrite_inline (flow_report_main_t * frm, + flow_report_t * fr, + ip4_address_t * collector_address, + ip4_address_t * src_address, + u16 collector_port, int variant) +{ + ip4_header_t *ip; + udp_header_t *udp; + ipfix_message_header_t *h; + ipfix_set_header_t *s; + ipfix_template_header_t *t; + ipfix_field_specifier_t *f; + ipfix_field_specifier_t *first_field; + u8 *rewrite = 0; + ip4_ipfix_template_packet_t *tp; + u32 field_count = 0; + flow_report_stream_t *stream; + flowperpkt_main_t *fm = &flowperpkt_main; + + stream = &frm->streams[fr->stream_index]; + + if (variant == FLOW_VARIANT_IPV4) + { + /* + * ip4 Supported Fields: + * + * ingressInterface, TLV type 10, u32 + * egressInterface, TLV type 14, u32 + * sourceIpv4Address, TLV type 8, u32 + * destinationIPv4Address, TLV type 12, u32 + * ipClassOfService, TLV type 5, u8 + * flowStartNanoseconds, TLV type 156, dateTimeNanoseconds (f64) + * Implementation: f64 nanoseconds since VPP started + * warning: wireshark doesn't really understand this TLV + * dataLinkFrameSize, TLV type 312, u16 + * warning: wireshark doesn't understand this TLV at all + */ + + /* Currently 7 fields */ + field_count += 7; + + /* allocate rewrite space */ + vec_validate_aligned + (rewrite, + sizeof (ip4_ipfix_template_packet_t) + + field_count * sizeof (ipfix_field_specifier_t) - 1, + CLIB_CACHE_LINE_BYTES); + } + else if (variant == FLOW_VARIANT_L2) + { + /* + * L2 Supported Fields: + * + * ingressInterface, TLV type 10, u32 + * egressInterface, TLV type 14, u32 + * sourceMacAddress, TLV type 56, u8[6] we hope + * destinationMacAddress, TLV type 57, u8[6] we hope + * ethernetType, TLV type 256, u16 + * flowStartNanoseconds, TLV type 156, dateTimeNanoseconds (f64) + * Implementation: f64 nanoseconds since VPP started + * warning: wireshark doesn't really understand this TLV + * dataLinkFrameSize, TLV type 312, u16 + * warning: wireshark doesn't understand this TLV at all + */ + + /* Currently 7 fields */ + field_count += 7; + + /* allocate rewrite space */ + vec_validate_aligned + (rewrite, + sizeof (ip4_ipfix_template_packet_t) + + field_count * sizeof (ipfix_field_specifier_t) - 1, + CLIB_CACHE_LINE_BYTES); + } + + tp = (ip4_ipfix_template_packet_t *) rewrite; + ip = (ip4_header_t *) & tp->ip4; + udp = (udp_header_t *) (ip + 1); + h = (ipfix_message_header_t *) (udp + 1); + s = (ipfix_set_header_t *) (h + 1); + t = (ipfix_template_header_t *) (s + 1); + first_field = f = (ipfix_field_specifier_t *) (t + 1); + + ip->ip_version_and_header_length = 0x45; + ip->ttl = 254; + ip->protocol = IP_PROTOCOL_UDP; + ip->src_address.as_u32 = src_address->as_u32; + ip->dst_address.as_u32 = collector_address->as_u32; + udp->src_port = clib_host_to_net_u16 (stream->src_port); + udp->dst_port = clib_host_to_net_u16 (collector_port); + udp->length = clib_host_to_net_u16 (vec_len (rewrite) - sizeof (*ip)); + + /* FIXUP: message header export_time */ + /* FIXUP: message header sequence_number */ + h->domain_id = clib_host_to_net_u32 (stream->domain_id); + + /* Add TLVs to the template */ + if (variant == FLOW_VARIANT_IPV4) + { + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , ingressInterface, + 4); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , egressInterface, + 4); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , sourceIPv4Address, + 4); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , destinationIPv4Address, 4); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , ipClassOfService, + 1); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , flowStartNanoseconds, + 8); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , dataLinkFrameSize, + 2); + f++; + } + else if (variant == FLOW_VARIANT_L2) + { + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , ingressInterface, + 4); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , egressInterface, + 4); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , sourceMacAddress, + 6); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , destinationMacAddress, 6); + f++; + f->e_id_length = ipfix_e_id_length (0 /* enterprise */ , ethernetType, + 2); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , flowStartNanoseconds, + 8); + f++; + f->e_id_length = + ipfix_e_id_length (0 /* enterprise */ , dataLinkFrameSize, + 2); + f++; + } + + /* Extend in the obvious way, right here... */ + + /* Back to the template packet... */ + ip = (ip4_header_t *) & tp->ip4; + udp = (udp_header_t *) (ip + 1); + + ASSERT (f - first_field); + /* Field count in this template */ + t->id_count = ipfix_id_count (fr->template_id, f - first_field); + + if (variant == FLOW_VARIANT_IPV4) + fm->ipv4_report_id = fr->template_id; + else if (variant == FLOW_VARIANT_L2) + fm->l2_report_id = fr->template_id; + + /* set length in octets */ + s->set_id_length = + ipfix_set_id_length (2 /* set_id */ , (u8 *) f - (u8 *) s); + + /* message length in octets */ + h->version_length = version_length ((u8 *) f - (u8 *) h); + + ip->length = clib_host_to_net_u16 ((u8 *) f - (u8 *) ip); + ip->checksum = ip4_header_checksum (ip); + + return rewrite; +} + +u8 * +flowperpkt_template_rewrite_ipv4 (flow_report_main_t * frm, + flow_report_t * fr, + ip4_address_t * collector_address, + ip4_address_t * src_address, + u16 collector_port) +{ + return flowperpkt_template_rewrite_inline + (frm, fr, collector_address, src_address, collector_port, + FLOW_VARIANT_IPV4); +} + +u8 * +flowperpkt_template_rewrite_l2 (flow_report_main_t * frm, + flow_report_t * fr, + ip4_address_t * collector_address, + ip4_address_t * src_address, + u16 collector_port) +{ + return flowperpkt_template_rewrite_inline + (frm, fr, collector_address, src_address, collector_port, + FLOW_VARIANT_L2); +} + + +/** + * @brief Flush accumulated data + * @param frm flow_report_main_t * + * @param fr flow_report_t * + * @param f vlib_frame_t * + * + * <em>Notes:</em> + * This function must simply return the incoming frame, or no template packets + * will be sent. + */ +vlib_frame_t * +flowperpkt_data_callback_ipv4 (flow_report_main_t * frm, + flow_report_t * fr, + vlib_frame_t * f, u32 * to_next, + u32 node_index) +{ + flowperpkt_flush_callback_ipv4 (); + return f; +} + +vlib_frame_t * +flowperpkt_data_callback_l2 (flow_report_main_t * frm, + flow_report_t * fr, + vlib_frame_t * f, u32 * to_next, u32 node_index) +{ + flowperpkt_flush_callback_l2 (); + return f; +} + +/** + * @brief configure / deconfigure the IPFIX flow-per-packet + * @param fm flowperpkt_main_t * fm + * @param sw_if_index u32 the desired interface + * @param is_add int 1 to enable the feature, 0 to disable it + * @returns 0 if successful, non-zero otherwise + */ + +static int flowperpkt_tx_interface_add_del_feature + (flowperpkt_main_t * fm, u32 sw_if_index, int which, int is_add) +{ + flow_report_main_t *frm = &flow_report_main; + vnet_flow_report_add_del_args_t _a, *a = &_a; + int rv; + + if (which == FLOW_VARIANT_IPV4 && !fm->ipv4_report_created) + { + memset (a, 0, sizeof (*a)); + a->rewrite_callback = flowperpkt_template_rewrite_ipv4; + a->flow_data_callback = flowperpkt_data_callback_ipv4; + a->is_add = 1; + a->domain_id = 1; /*$$$$ config parameter */ + a->src_port = 4739; /*$$$$ config parameter */ + fm->ipv4_report_created = 1; + + rv = vnet_flow_report_add_del (frm, a); + if (rv) + { + clib_warning ("vnet_flow_report_add_del returned %d", rv); + return -1; + } + } + else if (which == FLOW_VARIANT_L2 && !fm->l2_report_created) + { + memset (a, 0, sizeof (*a)); + a->rewrite_callback = flowperpkt_template_rewrite_l2; + a->flow_data_callback = flowperpkt_data_callback_l2; + a->is_add = 1; + a->domain_id = 1; /*$$$$ config parameter */ + a->src_port = 4739; /*$$$$ config parameter */ + fm->l2_report_created = 1; + + rv = vnet_flow_report_add_del (frm, a); + if (rv) + { + clib_warning ("vnet_flow_report_add_del returned %d", rv); + return -1; + } + } + + if (which == FLOW_VARIANT_IPV4) + vnet_feature_enable_disable ("ip4-output", "flowperpkt-ipv4", + sw_if_index, is_add, 0, 0); + else if (which == FLOW_VARIANT_L2) + vnet_feature_enable_disable ("interface-output", "flowperpkt-l2", + sw_if_index, is_add, 0, 0); + + return 0; +} + +/** + * @brief API message handler + * @param mp vl_api_flowperpkt_tx_interface_add_del_t * mp the api message + */ +void vl_api_flowperpkt_tx_interface_add_del_t_handler + (vl_api_flowperpkt_tx_interface_add_del_t * mp) +{ + flowperpkt_main_t *fm = &flowperpkt_main; + vl_api_flowperpkt_tx_interface_add_del_reply_t *rmp; + u32 sw_if_index = ntohl (mp->sw_if_index); + int rv = 0; + + VALIDATE_SW_IF_INDEX (mp); + + if (mp->which != FLOW_VARIANT_IPV4 && mp->which != FLOW_VARIANT_L2) + { + rv = VNET_API_ERROR_UNIMPLEMENTED; + goto out; + } + + rv = flowperpkt_tx_interface_add_del_feature (fm, sw_if_index, mp->which, + mp->is_add); +out: + BAD_SW_IF_INDEX_LABEL; + + REPLY_MACRO (VL_API_FLOWPERPKT_TX_INTERFACE_ADD_DEL_REPLY); +} + +/** + * @brief API message custom-dump function + * @param mp vl_api_flowperpkt_tx_interface_add_del_t * mp the api message + * @param handle void * print function handle + * @returns u8 * output string + */ +static void *vl_api_flowperpkt_tx_interface_add_del_t_print + (vl_api_flowperpkt_tx_interface_add_del_t * mp, void *handle) +{ + u8 *s; + + s = format (0, "SCRIPT: flowperpkt_tx_interface_add_del "); + s = format (s, "sw_if_index %d is_add %d which %d ", + clib_host_to_net_u32 (mp->sw_if_index), + (int) mp->is_add, (int) mp->which); + FINISH; +} + +/* List of message types that this plugin understands */ +#define foreach_flowperpkt_plugin_api_msg \ +_(FLOWPERPKT_TX_INTERFACE_ADD_DEL, flowperpkt_tx_interface_add_del) + +/** + * @brief plugin-api required function + * @param vm vlib_main_t * vlib main data structure pointer + * @param h vlib_plugin_handoff_t * handoff structure + * @param from_early_init int notused + * + * <em>Notes:</em> + * This routine exists to convince the vlib plugin framework that + * we haven't accidentally copied a random .dll into the plugin directory. + * + * Also collects global variable pointers passed from the vpp engine + */ +clib_error_t * +vlib_plugin_register (vlib_main_t * vm, vnet_plugin_handoff_t * h, + int from_early_init) +{ + flowperpkt_main_t *fm = &flowperpkt_main; + clib_error_t *error = 0; + + fm->vlib_main = vm; + fm->vnet_main = h->vnet_main; + + return error; +} + +static clib_error_t * +flowperpkt_tx_interface_add_del_feature_command_fn (vlib_main_t * vm, + unformat_input_t * input, + vlib_cli_command_t * cmd) +{ + flowperpkt_main_t *fm = &flowperpkt_main; + u32 sw_if_index = ~0; + int is_add = 1; + u8 which = FLOW_VARIANT_IPV4; + + int rv; + + while (unformat_check_input (input) != UNFORMAT_END_OF_INPUT) + { + if (unformat (input, "disable")) + is_add = 0; + else if (unformat (input, "%U", unformat_vnet_sw_interface, + fm->vnet_main, &sw_if_index)); + else if (unformat (input, "l2")) + which = FLOW_VARIANT_L2; + else + break; + } + + if (sw_if_index == ~0) + return clib_error_return (0, "Please specify an interface..."); + + rv = + flowperpkt_tx_interface_add_del_feature (fm, sw_if_index, which, is_add); + switch (rv) + { + case 0: + break; + + case VNET_API_ERROR_INVALID_SW_IF_INDEX: + return clib_error_return + (0, "Invalid interface, only works on physical ports"); + break; + + case VNET_API_ERROR_UNIMPLEMENTED: + return clib_error_return (0, "ip6 not supported"); + break; + + default: + return clib_error_return (0, "flowperpkt_enable_disable returned %d", + rv); + } + return 0; +} + +/*? + * '<em>flowperpkt feature add-del</em>' commands to enable/disable + * per-packet IPFIX flow record generation on an interface + * + * @cliexpar + * @parblock + * To enable per-packet IPFIX flow-record generation on an interface: + * @cliexcmd{flowperpkt feature add-del GigabitEthernet2/0/0} + * + * To disable per-packet IPFIX flow-record generation on an interface: + * @cliexcmd{flowperpkt feature add-del GigabitEthernet2/0/0 disable} + * @cliexend + * @endparblock +?*/ +/* *INDENT-OFF* */ +VLIB_CLI_COMMAND (flowperpkt_enable_disable_command, static) = { + .path = "flowperpkt feature add-del", + .short_help = + "flowperpkt feature add-del <interface-name> [disable]", + .function = flowperpkt_tx_interface_add_del_feature_command_fn, +}; +/* *INDENT-ON* */ + +/** + * @brief Set up the API message handling tables + * @param vm vlib_main_t * vlib main data structure pointer + * @returns 0 to indicate all is well + */ +static clib_error_t * +flowperpkt_plugin_api_hookup (vlib_main_t * vm) +{ + flowperpkt_main_t *fm = &flowperpkt_main; +#define _(N,n) \ + vl_msg_api_set_handlers((VL_API_##N + fm->msg_id_base), \ + #n, \ + vl_api_##n##_t_handler, \ + vl_noop_handler, \ + vl_api_##n##_t_endian, \ + vl_api_##n##_t_print, \ + sizeof(vl_api_##n##_t), 1); + foreach_flowperpkt_plugin_api_msg; +#undef _ + + return 0; +} + +#define vl_msg_name_crc_list +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_msg_name_crc_list + +static void +setup_message_id_table (flowperpkt_main_t * fm, api_main_t * am) +{ +#define _(id,n,crc) \ + vl_msg_api_add_msg_name_crc (am, #n "_" #crc, id + fm->msg_id_base); + foreach_vl_msg_name_crc_flowperpkt; +#undef _ +} + +/** + * @brief Set up the API message handling tables + * @param vm vlib_main_t * vlib main data structure pointer + * @returns 0 to indicate all is well, or a clib_error_t + */ +static clib_error_t * +flowperpkt_init (vlib_main_t * vm) +{ + flowperpkt_main_t *fm = &flowperpkt_main; + vlib_thread_main_t *tm = &vlib_thread_main; + clib_error_t *error = 0; + u32 num_threads; + u8 *name; + + /* Construct the API name */ + name = format (0, "flowperpkt_%08x%c", api_version, 0); + + /* Ask for a correctly-sized block of API message decode slots */ + fm->msg_id_base = vl_msg_api_get_msg_ids + ((char *) name, VL_MSG_FIRST_AVAILABLE); + + /* Hook up message handlers */ + error = flowperpkt_plugin_api_hookup (vm); + + /* Add our API messages to the global name_crc hash table */ + setup_message_id_table (fm, &api_main); + + vec_free (name); + + /* Decide how many worker threads we have */ + num_threads = 1 /* main thread */ + tm->n_eal_threads; + + /* Allocate per worker thread vectors */ + vec_validate (fm->ipv4_buffers_per_worker, num_threads - 1); + vec_validate (fm->l2_buffers_per_worker, num_threads - 1); + vec_validate (fm->ipv4_frames_per_worker, num_threads - 1); + vec_validate (fm->l2_frames_per_worker, num_threads - 1); + vec_validate (fm->ipv4_next_record_offset_per_worker, num_threads - 1); + vec_validate (fm->l2_next_record_offset_per_worker, num_threads - 1); + + /* Set up time reference pair */ + fm->vlib_time_0 = vlib_time_now (vm); + fm->nanosecond_time_0 = unix_time_now_nsec (); + + return error; +} + +VLIB_INIT_FUNCTION (flowperpkt_init); + +/* + * fd.io coding-style-patch-verification: ON + * + * Local Variables: + * eval: (c-set-style "gnu") + * End: + */ diff --git a/src/plugins/flowperpkt/flowperpkt.h b/src/plugins/flowperpkt/flowperpkt.h new file mode 100644 index 00000000000..20f6939dda5 --- /dev/null +++ b/src/plugins/flowperpkt/flowperpkt.h @@ -0,0 +1,90 @@ +/* + * flowperpkt.h - skeleton vpp engine plug-in header file + * + * Copyright (c) <current-year> <your-organization> + * 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. + */ +#ifndef __included_flowperpkt_h__ +#define __included_flowperpkt_h__ + +#include <vnet/vnet.h> +#include <vnet/ip/ip.h> +#include <vnet/ethernet/ethernet.h> + +#include <vppinfra/hash.h> +#include <vppinfra/error.h> +#include <vnet/flow/flow_report.h> +#include <vnet/flow/flow_report_classify.h> + +/** + * @file + * @brief flow-per-packet plugin header file + */ +typedef struct +{ + /** API message ID base */ + u16 msg_id_base; + + /** Have the reports [templates] been created? */ + int ipv4_report_created; + int l2_report_created; + + /** stream/template IDs */ + u16 ipv4_report_id; + u16 l2_report_id; + + /** ipfix buffers under construction, per-worker thread */ + vlib_buffer_t **ipv4_buffers_per_worker; + vlib_buffer_t **l2_buffers_per_worker; + + /** frames containing ipfix buffers, per-worker thread */ + vlib_frame_t **ipv4_frames_per_worker; + vlib_frame_t **l2_frames_per_worker; + + /** next record offset, per worker thread */ + u16 *ipv4_next_record_offset_per_worker; + u16 *l2_next_record_offset_per_worker; + + /** Time reference pair */ + u64 nanosecond_time_0; + f64 vlib_time_0; + + /** convenience vlib_main_t pointer */ + vlib_main_t *vlib_main; + /** convenience vnet_main_t pointer */ + vnet_main_t *vnet_main; +} flowperpkt_main_t; + +typedef enum +{ + FLOW_VARIANT_IPV4, + FLOW_VARIANT_L2, + FLOW_N_VARIANTS, +} flowperpkt_variant_t; + +extern flowperpkt_main_t flowperpkt_main; + +extern vlib_node_registration_t flowperpkt_ipv4_node; + +void flowperpkt_flush_callback_ipv4 (void); +void flowperpkt_flush_callback_l2 (void); + +#endif /* __included_flowperpkt_h__ */ + +/* + * fd.io coding-style-patch-verification: ON + * + * Local Variables: + * eval: (c-set-style "gnu") + * End: + */ diff --git a/src/plugins/flowperpkt/flowperpkt_all_api_h.h b/src/plugins/flowperpkt/flowperpkt_all_api_h.h new file mode 100644 index 00000000000..329c375abca --- /dev/null +++ b/src/plugins/flowperpkt/flowperpkt_all_api_h.h @@ -0,0 +1,18 @@ +/* + * flowperpkt_all_api_h.h - plug-in api #include file + * + * Copyright (c) <current-year> <your-organization> + * 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 the generated file, see BUILT_SOURCES in Makefile.am */ +#include <flowperpkt/flowperpkt.api.h> diff --git a/src/plugins/flowperpkt/flowperpkt_msg_enum.h b/src/plugins/flowperpkt/flowperpkt_msg_enum.h new file mode 100644 index 00000000000..3177e77a63b --- /dev/null +++ b/src/plugins/flowperpkt/flowperpkt_msg_enum.h @@ -0,0 +1,31 @@ +/* + * flowperpkt_msg_enum.h - vpp engine plug-in message enumeration + * + * Copyright (c) <current-year> <your-organization> + * 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. + */ +#ifndef included_flowperpkt_msg_enum_h +#define included_flowperpkt_msg_enum_h + +#include <vppinfra/byte_order.h> + +#define vl_msg_id(n,h) n, +typedef enum +{ +#include <flowperpkt/flowperpkt_all_api_h.h> + /* We'll want to know how many messages IDs we need... */ + VL_MSG_FIRST_AVAILABLE, +} vl_msg_id_t; +#undef vl_msg_id + +#endif /* included_flowperpkt_msg_enum_h */ diff --git a/src/plugins/flowperpkt/flowperpkt_plugin_doc.md b/src/plugins/flowperpkt/flowperpkt_plugin_doc.md new file mode 100644 index 00000000000..ed76c45c2dc --- /dev/null +++ b/src/plugins/flowperpkt/flowperpkt_plugin_doc.md @@ -0,0 +1,13 @@ +Per-packet IPFIX flow record plugin {#flowperpkt_plugin_doc} +=================================== + +## Introduction + +This plugin generates one ipfix record entry per packet transmitted +on interfaces which have the feature enabled + +## Sample configuration + +set ipfix exporter collector 192.168.6.2 src 192.168.6.1 template-interval 20 port 4739 path-mtu 1500 + +flowperpkt feature add-del GigabitEthernet2/3/0 diff --git a/src/plugins/flowperpkt/flowperpkt_test.c b/src/plugins/flowperpkt/flowperpkt_test.c new file mode 100644 index 00000000000..716818ffe0a --- /dev/null +++ b/src/plugins/flowperpkt/flowperpkt_test.c @@ -0,0 +1,234 @@ +/* + * flowperpkt.c - skeleton vpp-api-test plug-in + * + * Copyright (c) <current-year> <your-organization> + * 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 <vat/vat.h> +#include <vlibapi/api.h> +#include <vlibmemory/api.h> +#include <vlibsocket/api.h> +#include <vppinfra/error.h> + +/** + * @file vpp_api_test plugin + */ + +uword unformat_sw_if_index (unformat_input_t * input, va_list * args); + +/* Declare message IDs */ +#include <flowperpkt/flowperpkt_msg_enum.h> + +/* define message structures */ +#define vl_typedefs +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_typedefs + +/* declare message handlers for each api */ + +#define vl_endianfun /* define message structures */ +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_endianfun + +/* instantiate all the print functions we know about */ +#define vl_print(handle, ...) +#define vl_printfun +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_printfun + +/* Get the API version number. */ +#define vl_api_version(n,v) static u32 api_version=(v); +#include <flowperpkt/flowperpkt_all_api_h.h> +#undef vl_api_version + +typedef struct +{ + /** API message ID base */ + u16 msg_id_base; + /** vat_main_t pointer */ + vat_main_t *vat_main; +} flowperpkt_test_main_t; + +flowperpkt_test_main_t flowperpkt_test_main; + +#define foreach_standard_reply_retval_handler \ +_(flowperpkt_tx_interface_add_del_reply) + +#define _(n) \ + static void vl_api_##n##_t_handler \ + (vl_api_##n##_t * mp) \ + { \ + vat_main_t * vam = flowperpkt_test_main.vat_main; \ + i32 retval = ntohl(mp->retval); \ + if (vam->async_mode) { \ + vam->async_errors += (retval < 0); \ + } else { \ + vam->retval = retval; \ + vam->result_ready = 1; \ + } \ + } +foreach_standard_reply_retval_handler; +#undef _ + +/* + * Table of message reply handlers, must include boilerplate handlers + * we just generated + */ +#define foreach_vpe_api_reply_msg \ +_(FLOWPERPKT_TX_INTERFACE_ADD_DEL_REPLY, \ + flowperpkt_tx_interface_add_del_reply) + + +/* M: construct, but don't yet send a message */ + +#define M(T,t) \ +do { \ + vam->result_ready = 0; \ + mp = vl_msg_api_alloc(sizeof(*mp)); \ + memset (mp, 0, sizeof (*mp)); \ + mp->_vl_msg_id = ntohs (VL_API_##T + sm->msg_id_base); \ + mp->client_index = vam->my_client_index; \ +} while(0); + +#define M2(T,t,n) \ +do { \ + vam->result_ready = 0; \ + mp = vl_msg_api_alloc(sizeof(*mp)+(n)); \ + memset (mp, 0, sizeof (*mp)); \ + mp->_vl_msg_id = ntohs (VL_API_##T + sm->msg_id_base); \ + mp->client_index = vam->my_client_index; \ +} while(0); + +/* S: send a message */ +#define S (vl_msg_api_send_shmem (vam->vl_input_queue, (u8 *)&mp)) + +/* W: wait for results, with timeout */ +#define W \ +do { \ + timeout = vat_time_now (vam) + 1.0; \ + \ + while (vat_time_now (vam) < timeout) { \ + if (vam->result_ready == 1) { \ + return (vam->retval); \ + } \ + } \ + return -99; \ +} while(0); + +static int +api_flowperpkt_tx_interface_add_del (vat_main_t * vam) +{ + flowperpkt_test_main_t *sm = &flowperpkt_test_main; + unformat_input_t *i = vam->input; + f64 timeout; + int enable_disable = 1; + u8 which = 0; /* ipv4 by default */ + u32 sw_if_index = ~0; + vl_api_flowperpkt_tx_interface_add_del_t *mp; + + /* Parse args required to build the message */ + while (unformat_check_input (i) != UNFORMAT_END_OF_INPUT) + { + if (unformat (i, "%U", unformat_sw_if_index, vam, &sw_if_index)) + ; + else if (unformat (i, "sw_if_index %d", &sw_if_index)) + ; + else if (unformat (i, "disable")) + enable_disable = 0; + else if (unformat (i, "l2")) + which = 1; + else + break; + } + + if (sw_if_index == ~0) + { + errmsg ("missing interface name / explicit sw_if_index number \n"); + return -99; + } + + /* Construct the API message */ + M (FLOWPERPKT_TX_INTERFACE_ADD_DEL, flowperpkt_tx_interface_add_del); + mp->sw_if_index = ntohl (sw_if_index); + mp->is_add = enable_disable; + mp->which = which; + + /* send it... */ + S; + + /* Wait for a reply... */ + W; +} + +/* + * List of messages that the api test plugin sends, + * and that the data plane plugin processes + */ +#define foreach_vpe_api_msg \ +_(flowperpkt_tx_interface_add_del, "<intfc> [disable]") + +void +vat_api_hookup (vat_main_t * vam) +{ + flowperpkt_test_main_t *sm = &flowperpkt_test_main; + /* Hook up handlers for replies from the data plane plug-in */ +#define _(N,n) \ + vl_msg_api_set_handlers((VL_API_##N + sm->msg_id_base), \ + #n, \ + vl_api_##n##_t_handler, \ + vl_noop_handler, \ + vl_api_##n##_t_endian, \ + vl_api_##n##_t_print, \ + sizeof(vl_api_##n##_t), 1); + foreach_vpe_api_reply_msg; +#undef _ + + /* API messages we can send */ +#define _(n,h) hash_set_mem (vam->function_by_name, #n, api_##n); + foreach_vpe_api_msg; +#undef _ + + /* Help strings */ +#define _(n,h) hash_set_mem (vam->help_by_name, #n, h); + foreach_vpe_api_msg; +#undef _ +} + +clib_error_t * +vat_plugin_register (vat_main_t * vam) +{ + flowperpkt_test_main_t *sm = &flowperpkt_test_main; + u8 *name; + + sm->vat_main = vam; + + /* Ask the vpp engine for the first assigned message-id */ + name = format (0, "flowperpkt_%08x%c", api_version, 0); + sm->msg_id_base = vl_client_get_first_plugin_msg_id ((char *) name); + + /* Don't attempt to hook up API messages if the data plane plugin is AWOL */ + if (sm->msg_id_base != (u16) ~ 0) + vat_api_hookup (vam); + + vec_free (name); + + return 0; +} + +/* + * fd.io coding-style-patch-verification: ON + * + * Local Variables: + * eval: (c-set-style "gnu") + * End: + */ diff --git a/src/plugins/flowperpkt/l2_node.c b/src/plugins/flowperpkt/l2_node.c new file mode 100644 index 00000000000..1c2f681e1e1 --- /dev/null +++ b/src/plugins/flowperpkt/l2_node.c @@ -0,0 +1,561 @@ +/* + * l2_node.c - l2 ipfix-per-packet graph node + * + * Copyright (c) <current-year> <your-organization> + * 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 <vppinfra/error.h> +#include <flowperpkt/flowperpkt.h> + +/** + * @file l2 flow record generator graph node + */ + +typedef struct +{ + /** interface handle */ + u32 rx_sw_if_index; + u32 tx_sw_if_index; + /** src and dst L2 addresses */ + u8 src_mac[6]; + u8 dst_mac[6]; + /** Ethertype */ + u16 ethertype; + /** packet timestamp */ + u64 timestamp; + /** size of the buffer */ + u16 buffer_size; +} flowperpkt_l2_trace_t; + +/* packet trace format function */ +static u8 * +format_flowperpkt_l2_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 *); + flowperpkt_l2_trace_t *t = va_arg (*args, flowperpkt_l2_trace_t *); + + s = format (s, + "FLOWPERPKT-L2: rx_sw_if_index %d, tx_sw_if_index %d, src %U dst %U ethertype %0x2, timestamp %lld, size %d", + t->rx_sw_if_index, t->tx_sw_if_index, + format_ethernet_address, &t->src_mac, + format_ethernet_address, &t->dst_mac, + t->ethertype, t->timestamp, t->buffer_size); + return s; +} + +vlib_node_registration_t flowperpkt_l2_node; + +/* No counters at the moment */ +#define foreach_flowperpkt_l2_error + +typedef enum +{ +#define _(sym,str) FLOWPERPKT_ERROR_##sym, + foreach_flowperpkt_l2_error +#undef _ + FLOWPERPKT_N_ERROR, +} flowperpkt_l2_error_t; + +static char *flowperpkt_l2_error_strings[] = { +#define _(sym,string) string, + foreach_flowperpkt_l2_error +#undef _ +}; + +typedef enum +{ + FLOWPERPKT_L2_NEXT_DROP, + FLOWPERPKT_L2_NEXT_IP4_LOOKUP, + FLOWPERPKT_L2_N_NEXT, +} flowperpkt_l2_next_t; + +/** + * @brief add an entry to the flow record under construction + * @param vm vlib_main_t * current worker thread main structure pointer + * @param fm flowperpkt_main_t * flow-per-packet main structure pointer + * @param sw_if_index u32 interface handle + * @param tos u8 ToS bits from the packet + * @param timestamp u64 timestamp, nanoseconds since 1/1/70 + * @param length u16 ip length of the packet + * @param do_flush int 1 = flush all cached records, 0 = construct a record + */ + +static inline void +add_to_flow_record_l2 (vlib_main_t * vm, + vlib_node_runtime_t * node, + flowperpkt_main_t * fm, + u32 rx_sw_if_index, u32 tx_sw_if_index, + u8 * src_mac, u8 * dst_mac, + u16 ethertype, u64 timestamp, u16 length, int do_flush) +{ + u32 my_cpu_number = vm->cpu_index; + flow_report_main_t *frm = &flow_report_main; + ip4_header_t *ip; + udp_header_t *udp; + ip4_ipfix_template_packet_t *tp; + ipfix_message_header_t *h; + ipfix_set_header_t *s; + vlib_frame_t *f; + vlib_buffer_t *b0; + u16 offset; + u32 bi0; + vlib_buffer_free_list_t *fl; + + /* Find or allocate a buffer */ + b0 = fm->l2_buffers_per_worker[my_cpu_number]; + + /* Need to allocate a buffer? */ + if (PREDICT_FALSE (b0 == 0)) + { + /* Nothing to flush */ + if (do_flush) + return; + + /* $$$$ drop counter? */ + if (vlib_buffer_alloc (vm, &bi0, 1) != 1) + return; + + /* Initialize the buffer */ + b0 = fm->l2_buffers_per_worker[my_cpu_number] = + vlib_get_buffer (vm, bi0); + fl = + vlib_buffer_get_free_list (vm, VLIB_BUFFER_DEFAULT_FREE_LIST_INDEX); + vlib_buffer_init_for_free_list (b0, fl); + VLIB_BUFFER_TRACE_TRAJECTORY_INIT (b0); + offset = 0; + } + else + { + /* use the current buffer */ + bi0 = vlib_get_buffer_index (vm, b0); + offset = fm->l2_next_record_offset_per_worker[my_cpu_number]; + } + + /* Find or allocate a frame */ + f = fm->l2_frames_per_worker[my_cpu_number]; + if (PREDICT_FALSE (f == 0)) + { + u32 *to_next; + f = vlib_get_frame_to_node (vm, ip4_lookup_node.index); + fm->l2_frames_per_worker[my_cpu_number] = f; + + /* Enqueue the buffer */ + to_next = vlib_frame_vector_args (f); + to_next[0] = bi0; + f->n_vectors = 1; + } + + /* Fresh packet, construct header */ + if (PREDICT_FALSE (offset == 0)) + { + flow_report_stream_t *stream; + + stream = &frm->streams[0]; + + b0->current_data = 0; + b0->current_length = sizeof (*ip) + sizeof (*udp) + sizeof (*h) + + sizeof (*s); + b0->flags |= (VLIB_BUFFER_TOTAL_LENGTH_VALID | VLIB_BUFFER_FLOW_REPORT); + vnet_buffer (b0)->sw_if_index[VLIB_RX] = 0; + vnet_buffer (b0)->sw_if_index[VLIB_TX] = frm->fib_index; + + tp = vlib_buffer_get_current (b0); + ip = (ip4_header_t *) & tp->ip4; + udp = (udp_header_t *) (ip + 1); + h = (ipfix_message_header_t *) (udp + 1); + s = (ipfix_set_header_t *) (h + 1); + + ip->ip_version_and_header_length = 0x45; + ip->ttl = 254; + ip->protocol = IP_PROTOCOL_UDP; + ip->flags_and_fragment_offset = 0; + ip->src_address.as_u32 = frm->src_address.as_u32; + ip->dst_address.as_u32 = frm->ipfix_collector.as_u32; + udp->src_port = clib_host_to_net_u16 (UDP_DST_PORT_ipfix); + udp->dst_port = clib_host_to_net_u16 (UDP_DST_PORT_ipfix); + udp->checksum = 0; + + /* FIXUP: message header export_time */ + h->export_time = (u32) + (((f64) frm->unix_time_0) + + (vlib_time_now (frm->vlib_main) - frm->vlib_time_0)); + h->export_time = clib_host_to_net_u32 (h->export_time); + h->domain_id = clib_host_to_net_u32 (stream->domain_id); + + /* FIXUP: message header sequence_number */ + h->sequence_number = stream->sequence_number++; + h->sequence_number = clib_host_to_net_u32 (h->sequence_number); + + offset = (u32) (((u8 *) (s + 1)) - (u8 *) tp); + } + + /* Add data, unless we're flushing stale data */ + if (PREDICT_TRUE (do_flush == 0)) + { + + /* Add data */ + /* Ingress interface */ + { + u32 ingress_interface = clib_host_to_net_u32 (rx_sw_if_index); + clib_memcpy (b0->data + offset, &ingress_interface, + sizeof (ingress_interface)); + offset += sizeof (ingress_interface); + } + /* Egress interface */ + { + u32 egress_interface = clib_host_to_net_u32 (tx_sw_if_index); + clib_memcpy (b0->data + offset, &egress_interface, + sizeof (egress_interface)); + offset += sizeof (egress_interface); + } + /* src mac address */ + { + clib_memcpy (b0->data + offset, src_mac, 6); + offset += 6; + } + /* dst mac address */ + { + clib_memcpy (b0->data + offset, dst_mac, 6); + offset += 6; + } + + /* ethertype */ + b0->data[offset++] = ethertype >> 8; + b0->data[offset++] = ethertype & 0xFF; + + /* Timestamp */ + clib_memcpy (b0->data + offset, ×tamp, sizeof (f64)); + offset += sizeof (f64); + + /* pkt size */ + { + u16 pkt_size = clib_host_to_net_u16 (length); + clib_memcpy (b0->data + offset, &pkt_size, sizeof (pkt_size)); + offset += sizeof (pkt_size); + } + + b0->current_length += + /* 2*sw_if_index + 2*mac + ethertype + timestamp + length = 32 */ + 2 * sizeof (u32) + 12 + sizeof (u16) + sizeof (f64) + sizeof (u16); + + } + /* Time to flush the buffer? */ + if (PREDICT_FALSE + (do_flush || (offset + 2 * sizeof (u32) + 12 + sizeof (u16) + + +sizeof (f64) + sizeof (u16)) > frm->path_mtu)) + { + tp = vlib_buffer_get_current (b0); + ip = (ip4_header_t *) & tp->ip4; + udp = (udp_header_t *) (ip + 1); + h = (ipfix_message_header_t *) (udp + 1); + s = (ipfix_set_header_t *) (h + 1); + + s->set_id_length = ipfix_set_id_length (fm->l2_report_id, + b0->current_length - + (sizeof (*ip) + sizeof (*udp) + + sizeof (*h))); + h->version_length = version_length (b0->current_length - + (sizeof (*ip) + sizeof (*udp))); + + ip->length = clib_host_to_net_u16 (b0->current_length); + + ip->checksum = ip4_header_checksum (ip); + udp->length = clib_host_to_net_u16 (b0->current_length - sizeof (*ip)); + + if (frm->udp_checksum) + { + /* RFC 7011 section 10.3.2. */ + udp->checksum = ip4_tcp_udp_compute_checksum (vm, b0, ip); + if (udp->checksum == 0) + udp->checksum = 0xffff; + } + + ASSERT (ip->checksum == ip4_header_checksum (ip)); + + if (PREDICT_FALSE (vlib_get_trace_count (vm, node) > 0)) + { + vlib_trace_buffer (vm, node, FLOWPERPKT_L2_NEXT_IP4_LOOKUP, b0, + 0 /* follow chain */ ); + flowperpkt_l2_trace_t *t = + vlib_add_trace (vm, node, b0, sizeof (*t)); + memset (t, 0, sizeof (*t)); + t->rx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_RX]; + t->tx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_TX]; + t->buffer_size = b0->current_length; + } + + vlib_put_frame_to_node (vm, ip4_lookup_node.index, + fm->l2_frames_per_worker[my_cpu_number]); + fm->l2_frames_per_worker[my_cpu_number] = 0; + fm->l2_buffers_per_worker[my_cpu_number] = 0; + offset = 0; + } + + fm->l2_next_record_offset_per_worker[my_cpu_number] = offset; +} + +void +flowperpkt_flush_callback_l2 (void) +{ + vlib_main_t *vm = vlib_get_main (); + flowperpkt_main_t *fm = &flowperpkt_main; + vlib_node_runtime_t *node; + node = vlib_node_get_runtime (vm, flowperpkt_l2_node.index); + + add_to_flow_record_l2 (vm, node, fm, 0 /* rx_sw_if_index */ , + 0 /* tx_sw_if_index */ , + 0 /* src mac */ , + 0 /* dst mac */ , + 0 /* ethertype */ , + 0ULL /* timestamp */ , + 0 /* length */ , + 1 /* do_flush */ ); +} + + +static uword +flowperpkt_l2_node_fn (vlib_main_t * vm, + vlib_node_runtime_t * node, vlib_frame_t * frame) +{ + u32 n_left_from, *from, *to_next; + flowperpkt_l2_next_t next_index; + flowperpkt_main_t *fm = &flowperpkt_main; + u64 now; + + now = (u64) ((vlib_time_now (vm) - fm->vlib_time_0) * 1e9); + now += fm->nanosecond_time_0; + + from = vlib_frame_vector_args (frame); + n_left_from = frame->n_vectors; + next_index = node->cached_next_index; + + while (n_left_from > 0) + { + u32 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) + { + u32 next0 = FLOWPERPKT_L2_NEXT_DROP; + u32 next1 = FLOWPERPKT_L2_NEXT_DROP; + ethernet_header_t *eh0, *eh1; + u16 len0, len1; + u32 bi0, bi1; + vlib_buffer_t *b0, *b1; + + /* 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, CLIB_CACHE_LINE_BYTES, STORE); + CLIB_PREFETCH (p3->data, CLIB_CACHE_LINE_BYTES, STORE); + } + + /* speculatively enqueue b0 and b1 to the current next frame */ + to_next[0] = bi0 = from[0]; + to_next[1] = bi1 = from[1]; + from += 2; + to_next += 2; + n_left_from -= 2; + n_left_to_next -= 2; + + b0 = vlib_get_buffer (vm, bi0); + b1 = vlib_get_buffer (vm, bi1); + + vnet_feature_next (vnet_buffer (b0)->sw_if_index[VLIB_TX], + &next0, b0); + vnet_feature_next (vnet_buffer (b1)->sw_if_index[VLIB_TX], + &next1, b1); + + eh0 = vlib_buffer_get_current (b0); + len0 = vlib_buffer_length_in_chain (vm, b0); + + if (PREDICT_TRUE ((b0->flags & VLIB_BUFFER_FLOW_REPORT) == 0)) + add_to_flow_record_l2 (vm, node, fm, + vnet_buffer (b0)->sw_if_index[VLIB_RX], + vnet_buffer (b0)->sw_if_index[VLIB_TX], + eh0->src_address, + eh0->dst_address, + eh0->type, now, len0, 0 /* flush */ ); + + eh1 = vlib_buffer_get_current (b0); + len1 = vlib_buffer_length_in_chain (vm, b0); + + if (PREDICT_TRUE ((b1->flags & VLIB_BUFFER_FLOW_REPORT) == 0)) + add_to_flow_record_l2 (vm, node, fm, + vnet_buffer (b1)->sw_if_index[VLIB_RX], + vnet_buffer (b1)->sw_if_index[VLIB_TX], + eh1->src_address, + eh1->dst_address, + eh1->type, now, len1, 0 /* flush */ ); + + if (PREDICT_FALSE ((node->flags & VLIB_NODE_FLAG_TRACE))) + { + if (b0->flags & VLIB_BUFFER_IS_TRACED) + { + flowperpkt_l2_trace_t *t = + vlib_add_trace (vm, node, b0, sizeof (*t)); + t->rx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_RX]; + t->tx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_TX]; + clib_memcpy (t->src_mac, eh0->src_address, 6); + clib_memcpy (t->dst_mac, eh0->dst_address, 6); + t->ethertype = clib_net_to_host_u16 (eh0->type); + t->timestamp = now; + t->buffer_size = len0; + } + if (b1->flags & VLIB_BUFFER_IS_TRACED) + { + flowperpkt_l2_trace_t *t = + vlib_add_trace (vm, node, b1, sizeof (*t)); + t->rx_sw_if_index = vnet_buffer (b1)->sw_if_index[VLIB_RX]; + t->tx_sw_if_index = vnet_buffer (b1)->sw_if_index[VLIB_TX]; + clib_memcpy (t->src_mac, eh1->src_address, 6); + clib_memcpy (t->dst_mac, eh1->dst_address, 6); + t->ethertype = clib_net_to_host_u16 (eh1->type); + t->timestamp = now; + t->buffer_size = len1; + } + } + + /* verify speculative enqueues, maybe switch current next frame */ + 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; + vlib_buffer_t *b0; + u32 next0 = FLOWPERPKT_L2_NEXT_DROP; + ethernet_header_t *eh0; + u16 len0; + + /* 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); + + vnet_feature_next (vnet_buffer (b0)->sw_if_index[VLIB_TX], + &next0, b0); + + eh0 = vlib_buffer_get_current (b0); + len0 = vlib_buffer_length_in_chain (vm, b0); + + if (PREDICT_TRUE ((b0->flags & VLIB_BUFFER_FLOW_REPORT) == 0)) + add_to_flow_record_l2 (vm, node, fm, + vnet_buffer (b0)->sw_if_index[VLIB_RX], + vnet_buffer (b0)->sw_if_index[VLIB_TX], + eh0->src_address, + eh0->dst_address, + eh0->type, now, len0, 0 /* flush */ ); + + if (PREDICT_FALSE ((node->flags & VLIB_NODE_FLAG_TRACE) + && (b0->flags & VLIB_BUFFER_IS_TRACED))) + { + flowperpkt_l2_trace_t *t = + vlib_add_trace (vm, node, b0, sizeof (*t)); + t->rx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_RX]; + t->tx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_TX]; + clib_memcpy (t->src_mac, eh0->src_address, 6); + clib_memcpy (t->dst_mac, eh0->dst_address, 6); + t->ethertype = clib_net_to_host_u16 (eh0->type); + t->timestamp = now; + t->buffer_size = len0; + } + + /* 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; +} + +/** + * @brief IPFIX l2 flow-per-packet graph node + * @node flowperpkt-l2 + * + * This is the IPFIX flow-record-per-packet node. + * + * @param vm vlib_main_t corresponding to the current thread. + * @param node vlib_node_runtime_t data for this node. + * @param frame vlib_frame_t whose contents should be dispatched. + * + * @par Graph mechanics: buffer metadata, next index usage + * + * <em>Uses:</em> + * - <code>vnet_buffer(b)->ip.save_rewrite_length</code> + * - tells the node the length of the rewrite which was applied in + * ip4/6_rewrite_inline, allows the code to find the IP header without + * having to parse L2 headers, or make stupid assumptions about their + * length. + * - <code>vnet_buffer(b)->flags & VLIB_BUFFER_FLOW_REPORT</code> + * - Used to suppress flow record generation for flow record packets. + * + * <em>Sets:</em> + * - <code>vnet_buffer(b)->flags & VLIB_BUFFER_FLOW_REPORT</code> + * - To suppress flow record generation for flow record packets + * + * <em>Next Index:</em> + * - Next configured output feature on the interface, usually + * "interface-output." Generated flow records head for ip4-lookup + */ + +/* *INDENT-OFF* */ +VLIB_REGISTER_NODE (flowperpkt_l2_node) = { + .function = flowperpkt_l2_node_fn, + .name = "flowperpkt-l2", + .vector_size = sizeof (u32), + .format_trace = format_flowperpkt_l2_trace, + .type = VLIB_NODE_TYPE_INTERNAL, + + .n_errors = ARRAY_LEN(flowperpkt_l2_error_strings), + .error_strings = flowperpkt_l2_error_strings, + + .n_next_nodes = FLOWPERPKT_L2_N_NEXT, + + /* edit / add dispositions here */ + .next_nodes = { + [FLOWPERPKT_L2_NEXT_DROP] = "error-drop", + [FLOWPERPKT_L2_NEXT_IP4_LOOKUP] = "ip4-lookup", + }, +}; +/* *INDENT-ON* */ + +/* + * fd.io coding-style-patch-verification: ON + * + * Local Variables: + * eval: (c-set-style "gnu") + * End: + */ diff --git a/src/plugins/flowperpkt/node.c b/src/plugins/flowperpkt/node.c new file mode 100644 index 00000000000..f77f087dc78 --- /dev/null +++ b/src/plugins/flowperpkt/node.c @@ -0,0 +1,574 @@ +/* + * node.c - ipv4 ipfix-per-packet graph node + * + * Copyright (c) <current-year> <your-organization> + * 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 <vppinfra/error.h> +#include <flowperpkt/flowperpkt.h> + +/** + * @file ipv4 flow record generator graph node + */ + +typedef struct +{ + /** interface handle */ + u32 rx_sw_if_index; + u32 tx_sw_if_index; + u32 src_address; + u32 dst_address; + /** ToS bits */ + u8 tos; + /** packet timestamp */ + u64 timestamp; + /** size of the buffer */ + u16 buffer_size; +} flowperpkt_ipv4_trace_t; + +/* packet trace format function */ +static u8 * +format_flowperpkt_ipv4_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 *); + flowperpkt_ipv4_trace_t *t = va_arg (*args, flowperpkt_ipv4_trace_t *); + + s = format (s, + "FLOWPERPKT-V4: rx_sw_if_index %d, tx_sw_if_index %d, src %U dst %U tos %0x2, timestamp %lld, size %d", + t->rx_sw_if_index, t->tx_sw_if_index, + format_ip4_address, &t->src_address, + format_ip4_address, &t->dst_address, + t->tos, t->timestamp, t->buffer_size); + return s; +} + +vlib_node_registration_t flowperpkt_ipv4_node; + +/* No counters at the moment */ +#define foreach_flowperpkt_ipv4_error + +typedef enum +{ +#define _(sym,str) FLOWPERPKT_ERROR_##sym, + foreach_flowperpkt_ipv4_error +#undef _ + FLOWPERPKT_N_ERROR, +} flowperpkt_ipv4_error_t; + +static char *flowperpkt_ipv4_error_strings[] = { +#define _(sym,string) string, + foreach_flowperpkt_ipv4_error +#undef _ +}; + +typedef enum +{ + FLOWPERPKT_IPV4_NEXT_DROP, + FLOWPERPKT_IPV4_NEXT_LOOKUP, + FLOWPERPKT_IPV4_N_NEXT, +} flowperpkt_ipv4_next_t; + +/** + * @brief add an entry to the flow record under construction + * @param vm vlib_main_t * current worker thread main structure pointer + * @param fm flowperpkt_main_t * flow-per-packet main structure pointer + * @param sw_if_index u32 interface handle + * @param tos u8 ToS bits from the packet + * @param timestamp u64 timestamp, nanoseconds since 1/1/70 + * @param length u16 ip length of the packet + * @param do_flush int 1 = flush all cached records, 0 = construct a record + */ + +static inline void +add_to_flow_record_ipv4 (vlib_main_t * vm, + vlib_node_runtime_t * node, + flowperpkt_main_t * fm, + u32 rx_sw_if_index, u32 tx_sw_if_index, + u32 src_address, u32 dst_address, + u8 tos, u64 timestamp, u16 length, int do_flush) +{ + u32 my_cpu_number = vm->cpu_index; + flow_report_main_t *frm = &flow_report_main; + ip4_header_t *ip; + udp_header_t *udp; + ip4_ipfix_template_packet_t *tp; + ipfix_message_header_t *h; + ipfix_set_header_t *s; + vlib_frame_t *f; + vlib_buffer_t *b0; + u16 offset; + u32 bi0; + vlib_buffer_free_list_t *fl; + + /* Find or allocate a buffer */ + b0 = fm->ipv4_buffers_per_worker[my_cpu_number]; + + /* Need to allocate a buffer? */ + if (PREDICT_FALSE (b0 == 0)) + { + /* Nothing to flush */ + if (do_flush) + return; + + /* $$$$ drop counter? */ + if (vlib_buffer_alloc (vm, &bi0, 1) != 1) + return; + + /* Initialize the buffer */ + b0 = fm->ipv4_buffers_per_worker[my_cpu_number] = + vlib_get_buffer (vm, bi0); + fl = + vlib_buffer_get_free_list (vm, VLIB_BUFFER_DEFAULT_FREE_LIST_INDEX); + vlib_buffer_init_for_free_list (b0, fl); + VLIB_BUFFER_TRACE_TRAJECTORY_INIT (b0); + offset = 0; + } + else + { + /* use the current buffer */ + bi0 = vlib_get_buffer_index (vm, b0); + offset = fm->ipv4_next_record_offset_per_worker[my_cpu_number]; + } + + /* Find or allocate a frame */ + f = fm->ipv4_frames_per_worker[my_cpu_number]; + if (PREDICT_FALSE (f == 0)) + { + u32 *to_next; + f = vlib_get_frame_to_node (vm, ip4_lookup_node.index); + fm->ipv4_frames_per_worker[my_cpu_number] = f; + + /* Enqueue the buffer */ + to_next = vlib_frame_vector_args (f); + to_next[0] = bi0; + f->n_vectors = 1; + } + + /* Fresh packet, construct header */ + if (PREDICT_FALSE (offset == 0)) + { + flow_report_stream_t *stream; + + stream = &frm->streams[0]; + + b0->current_data = 0; + b0->current_length = sizeof (*ip) + sizeof (*udp) + sizeof (*h) + + sizeof (*s); + b0->flags |= (VLIB_BUFFER_TOTAL_LENGTH_VALID | VLIB_BUFFER_FLOW_REPORT); + vnet_buffer (b0)->sw_if_index[VLIB_RX] = 0; + vnet_buffer (b0)->sw_if_index[VLIB_TX] = frm->fib_index; + + tp = vlib_buffer_get_current (b0); + ip = (ip4_header_t *) & tp->ip4; + udp = (udp_header_t *) (ip + 1); + h = (ipfix_message_header_t *) (udp + 1); + s = (ipfix_set_header_t *) (h + 1); + + ip->ip_version_and_header_length = 0x45; + ip->ttl = 254; + ip->protocol = IP_PROTOCOL_UDP; + ip->flags_and_fragment_offset = 0; + ip->src_address.as_u32 = frm->src_address.as_u32; + ip->dst_address.as_u32 = frm->ipfix_collector.as_u32; + udp->src_port = clib_host_to_net_u16 (UDP_DST_PORT_ipfix); + udp->dst_port = clib_host_to_net_u16 (UDP_DST_PORT_ipfix); + udp->checksum = 0; + + /* FIXUP: message header export_time */ + h->export_time = (u32) + (((f64) frm->unix_time_0) + + (vlib_time_now (frm->vlib_main) - frm->vlib_time_0)); + h->export_time = clib_host_to_net_u32 (h->export_time); + h->domain_id = clib_host_to_net_u32 (stream->domain_id); + + /* FIXUP: message header sequence_number */ + h->sequence_number = stream->sequence_number++; + h->sequence_number = clib_host_to_net_u32 (h->sequence_number); + + offset = (u32) (((u8 *) (s + 1)) - (u8 *) tp); + } + + /* Add data, unless we're flushing stale data */ + if (PREDICT_TRUE (do_flush == 0)) + { + + /* Add data */ + /* Ingress interface */ + { + u32 ingress_interface = clib_host_to_net_u32 (rx_sw_if_index); + clib_memcpy (b0->data + offset, &ingress_interface, + sizeof (ingress_interface)); + offset += sizeof (ingress_interface); + } + /* Egress interface */ + { + u32 egress_interface = clib_host_to_net_u32 (tx_sw_if_index); + clib_memcpy (b0->data + offset, &egress_interface, + sizeof (egress_interface)); + offset += sizeof (egress_interface); + } + /* ip4 src address */ + { + clib_memcpy (b0->data + offset, &src_address, sizeof (src_address)); + offset += sizeof (src_address); + } + /* ip4 dst address */ + { + clib_memcpy (b0->data + offset, &dst_address, sizeof (dst_address)); + offset += sizeof (dst_address); + } + + /* ToS */ + b0->data[offset++] = tos; + + /* Timestamp */ + clib_memcpy (b0->data + offset, ×tamp, sizeof (f64)); + offset += sizeof (f64); + + /* pkt size */ + { + u16 pkt_size = clib_host_to_net_u16 (length); + clib_memcpy (b0->data + offset, &pkt_size, sizeof (pkt_size)); + offset += sizeof (pkt_size); + } + + b0->current_length += + /* sw_if_index + tos + timestamp + length = 15 */ + 4 * sizeof (u32) + sizeof (u8) + sizeof (f64) + sizeof (u16); + + } + /* Time to flush the buffer? */ + if (PREDICT_FALSE + (do_flush || (offset + 4 * sizeof (u32) + sizeof (u8) + + sizeof (f64) + sizeof (u16)) > frm->path_mtu)) + { + tp = vlib_buffer_get_current (b0); + ip = (ip4_header_t *) & tp->ip4; + udp = (udp_header_t *) (ip + 1); + h = (ipfix_message_header_t *) (udp + 1); + s = (ipfix_set_header_t *) (h + 1); + + s->set_id_length = ipfix_set_id_length (fm->ipv4_report_id, + b0->current_length - + (sizeof (*ip) + sizeof (*udp) + + sizeof (*h))); + h->version_length = version_length (b0->current_length - + (sizeof (*ip) + sizeof (*udp))); + + ip->length = clib_host_to_net_u16 (b0->current_length); + + ip->checksum = ip4_header_checksum (ip); + udp->length = clib_host_to_net_u16 (b0->current_length - sizeof (*ip)); + + if (frm->udp_checksum) + { + /* RFC 7011 section 10.3.2. */ + udp->checksum = ip4_tcp_udp_compute_checksum (vm, b0, ip); + if (udp->checksum == 0) + udp->checksum = 0xffff; + } + + ASSERT (ip->checksum == ip4_header_checksum (ip)); + + if (PREDICT_FALSE (vlib_get_trace_count (vm, node) > 0)) + { + vlib_trace_buffer (vm, node, FLOWPERPKT_IPV4_NEXT_LOOKUP, b0, + 0 /* follow chain */ ); + flowperpkt_ipv4_trace_t *t = + vlib_add_trace (vm, node, b0, sizeof (*t)); + t->rx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_RX]; + t->tx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_TX]; + t->src_address = 0; + t->dst_address = 0; + t->tos = 0; + t->timestamp = 0; + t->buffer_size = b0->current_length; + } + + vlib_put_frame_to_node (vm, ip4_lookup_node.index, + fm->ipv4_frames_per_worker[my_cpu_number]); + fm->ipv4_frames_per_worker[my_cpu_number] = 0; + fm->ipv4_buffers_per_worker[my_cpu_number] = 0; + offset = 0; + } + + fm->ipv4_next_record_offset_per_worker[my_cpu_number] = offset; +} + +void +flowperpkt_flush_callback_ipv4 (void) +{ + vlib_main_t *vm = vlib_get_main (); + flowperpkt_main_t *fm = &flowperpkt_main; + vlib_node_runtime_t *node; + node = vlib_node_get_runtime (vm, flowperpkt_ipv4_node.index); + + add_to_flow_record_ipv4 (vm, node, fm, 0 /* rx_sw_if_index */ , + 0 /* tx_sw_if_index */ , + 0 /* src_address */ , + 0 /* dst_address */ , + 0 /* ToS */ , + 0ULL /* timestamp */ , + 0 /* length */ , + 1 /* do_flush */ ); +} + + +static uword +flowperpkt_ipv4_node_fn (vlib_main_t * vm, + vlib_node_runtime_t * node, vlib_frame_t * frame) +{ + u32 n_left_from, *from, *to_next; + flowperpkt_ipv4_next_t next_index; + flowperpkt_main_t *fm = &flowperpkt_main; + u64 now; + + now = (u64) ((vlib_time_now (vm) - fm->vlib_time_0) * 1e9); + now += fm->nanosecond_time_0; + + from = vlib_frame_vector_args (frame); + n_left_from = frame->n_vectors; + next_index = node->cached_next_index; + + while (n_left_from > 0) + { + u32 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) + { + u32 next0 = FLOWPERPKT_IPV4_NEXT_DROP; + u32 next1 = FLOWPERPKT_IPV4_NEXT_DROP; + ip4_header_t *ip0, *ip1; + u16 len0, len1; + u32 bi0, bi1; + vlib_buffer_t *b0, *b1; + + /* 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, CLIB_CACHE_LINE_BYTES, STORE); + CLIB_PREFETCH (p3->data, CLIB_CACHE_LINE_BYTES, STORE); + } + + /* speculatively enqueue b0 and b1 to the current next frame */ + to_next[0] = bi0 = from[0]; + to_next[1] = bi1 = from[1]; + from += 2; + to_next += 2; + n_left_from -= 2; + n_left_to_next -= 2; + + b0 = vlib_get_buffer (vm, bi0); + b1 = vlib_get_buffer (vm, bi1); + + vnet_feature_next (vnet_buffer (b0)->sw_if_index[VLIB_TX], + &next0, b0); + vnet_feature_next (vnet_buffer (b1)->sw_if_index[VLIB_TX], + &next1, b1); + + ip0 = (ip4_header_t *) ((u8 *) vlib_buffer_get_current (b0) + + vnet_buffer (b0)->ip.save_rewrite_length); + + len0 = vlib_buffer_length_in_chain (vm, b0); + + if (PREDICT_TRUE ((b0->flags & VLIB_BUFFER_FLOW_REPORT) == 0)) + add_to_flow_record_ipv4 (vm, node, fm, + vnet_buffer (b0)->sw_if_index[VLIB_RX], + vnet_buffer (b0)->sw_if_index[VLIB_TX], + ip0->src_address.as_u32, + ip0->dst_address.as_u32, + ip0->tos, now, len0, 0 /* flush */ ); + + ip1 = (ip4_header_t *) ((u8 *) vlib_buffer_get_current (b1) + + vnet_buffer (b1)->ip.save_rewrite_length); + len1 = vlib_buffer_length_in_chain (vm, b1); + + if (PREDICT_TRUE ((b1->flags & VLIB_BUFFER_FLOW_REPORT) == 0)) + add_to_flow_record_ipv4 (vm, node, fm, + vnet_buffer (b1)->sw_if_index[VLIB_RX], + vnet_buffer (b1)->sw_if_index[VLIB_TX], + ip1->src_address.as_u32, + ip1->dst_address.as_u32, + ip1->tos, now, len1, 0 /* flush */ ); + + if (PREDICT_FALSE ((node->flags & VLIB_NODE_FLAG_TRACE))) + { + if (b0->flags & VLIB_BUFFER_IS_TRACED) + { + flowperpkt_ipv4_trace_t *t = + vlib_add_trace (vm, node, b0, sizeof (*t)); + t->rx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_RX]; + t->tx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_TX]; + t->src_address = ip0->src_address.as_u32; + t->dst_address = ip0->dst_address.as_u32; + t->tos = ip0->tos; + t->timestamp = now; + t->buffer_size = len0; + } + if (b1->flags & VLIB_BUFFER_IS_TRACED) + { + flowperpkt_ipv4_trace_t *t = + vlib_add_trace (vm, node, b1, sizeof (*t)); + t->rx_sw_if_index = vnet_buffer (b1)->sw_if_index[VLIB_RX]; + t->tx_sw_if_index = vnet_buffer (b1)->sw_if_index[VLIB_TX]; + t->src_address = ip1->src_address.as_u32; + t->dst_address = ip1->dst_address.as_u32; + t->tos = ip1->tos; + t->timestamp = now; + t->buffer_size = len1; + } + } + + /* verify speculative enqueues, maybe switch current next frame */ + 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; + vlib_buffer_t *b0; + u32 next0 = FLOWPERPKT_IPV4_NEXT_DROP; + ip4_header_t *ip0; + u16 len0; + + /* 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); + + vnet_feature_next (vnet_buffer (b0)->sw_if_index[VLIB_TX], + &next0, b0); + + ip0 = (ip4_header_t *) ((u8 *) vlib_buffer_get_current (b0) + + vnet_buffer (b0)->ip.save_rewrite_length); + /* + * egressInterface, TLV type 14, u32 + * ipClassOfService, TLV type 5, u8 + * flowStartNanoseconds, TLV type 156, dateTimeNanoseconds (f64) + * Implementation: f64 nanoseconds since VPP started + * dataLinkFrameSize, TLV type 312, u16 + */ + len0 = vlib_buffer_length_in_chain (vm, b0); + + if (PREDICT_TRUE ((b0->flags & VLIB_BUFFER_FLOW_REPORT) == 0)) + add_to_flow_record_ipv4 (vm, node, fm, + vnet_buffer (b0)->sw_if_index[VLIB_RX], + vnet_buffer (b0)->sw_if_index[VLIB_TX], + ip0->src_address.as_u32, + ip0->dst_address.as_u32, + ip0->tos, now, len0, 0 /* flush */ ); + + if (PREDICT_FALSE ((node->flags & VLIB_NODE_FLAG_TRACE) + && (b0->flags & VLIB_BUFFER_IS_TRACED))) + { + flowperpkt_ipv4_trace_t *t = + vlib_add_trace (vm, node, b0, sizeof (*t)); + t->rx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_RX]; + t->tx_sw_if_index = vnet_buffer (b0)->sw_if_index[VLIB_TX]; + t->src_address = ip0->src_address.as_u32; + t->dst_address = ip0->dst_address.as_u32; + t->tos = ip0->tos; + t->timestamp = now; + t->buffer_size = len0; + } + + /* 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; +} + +/** + * @brief IPFIX ipv4 flow-per-packet graph node + * @node flowperpkt-ipv4 + * + * This is the IPFIX flow-record-per-packet node. + * + * @param vm vlib_main_t corresponding to the current thread. + * @param node vlib_node_runtime_t data for this node. + * @param frame vlib_frame_t whose contents should be dispatched. + * + * @par Graph mechanics: buffer metadata, next index usage + * + * <em>Uses:</em> + * - <code>vnet_buffer(b)->ip.save_rewrite_length</code> + * - tells the node the length of the rewrite which was applied in + * ip4/6_rewrite_inline, allows the code to find the IP header without + * having to parse L2 headers, or make stupid assumptions about their + * length. + * - <code>vnet_buffer(b)->flags & VLIB_BUFFER_FLOW_REPORT</code> + * - Used to suppress flow record generation for flow record packets. + * + * <em>Sets:</em> + * - <code>vnet_buffer(b)->flags & VLIB_BUFFER_FLOW_REPORT</code> + * - To suppress flow record generation for flow record packets + * + * <em>Next Index:</em> + * - Next configured output feature on the interface, usually + * "interface-output." Generated flow records head for ip4-lookup + */ + +/* *INDENT-OFF* */ +VLIB_REGISTER_NODE (flowperpkt_ipv4_node) = { + .function = flowperpkt_ipv4_node_fn, + .name = "flowperpkt-ipv4", + .vector_size = sizeof (u32), + .format_trace = format_flowperpkt_ipv4_trace, + .type = VLIB_NODE_TYPE_INTERNAL, + + .n_errors = ARRAY_LEN(flowperpkt_ipv4_error_strings), + .error_strings = flowperpkt_ipv4_error_strings, + + .n_next_nodes = FLOWPERPKT_IPV4_N_NEXT, + + /* edit / add dispositions here */ + .next_nodes = { + [FLOWPERPKT_IPV4_NEXT_DROP] = "error-drop", + /* Used only to trace ipfix data packets */ + [FLOWPERPKT_IPV4_NEXT_LOOKUP] = "ip4-lookup", + }, +}; +/* *INDENT-ON* */ + +/* + * fd.io coding-style-patch-verification: ON + * + * Local Variables: + * eval: (c-set-style "gnu") + * End: + */ diff --git a/src/plugins/ila.am b/src/plugins/ila.am new file mode 100644 index 00000000000..d900f3eb307 --- /dev/null +++ b/src/plugins/ila.am @@ -0,0 +1,20 @@ +# Copyright (c) 2016 Cisco Systems, Inc. +# 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. + +vppplugins_LTLIBRARIES += ila_plugin.la + +ila_plugin_la_SOURCES = ila/ila.c + +noinst_HEADERS += ila/ila.h + +# vi:syntax=automake diff --git a/src/plugins/ila/ila.c b/src/plugins/ila/ila.c new file mode 100644 index 00000000000..336f4cf560c --- /dev/null +++ b/src/plugins/ila/ila.c @@ -0,0 +1,1070 @@ +/* + * 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: + * + * 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 <ila/ila.h> +#include <vnet/plugin/plugin.h> +#include <vnet/ip/lookup.h> +#include <vnet/dpo/dpo.h> +#include <vnet/fib/fib_table.h> + +static ila_main_t ila_main; + +#define ILA_TABLE_DEFAULT_HASH_NUM_BUCKETS (64 * 1024) +#define ILA_TABLE_DEFAULT_HASH_MEMORY_SIZE (32<<20) + +#define foreach_ila_error \ + _(NONE, "valid ILA packets") + +typedef enum { +#define _(sym,str) ILA_ERROR_##sym, + foreach_ila_error +#undef _ + ILA_N_ERROR, +} ila_error_t; + +static char *ila_error_strings[] = { +#define _(sym,string) string, + foreach_ila_error +#undef _ +}; + +typedef enum { + ILA_ILA2SIR_NEXT_DROP, + ILA_ILA2SIR_N_NEXT, +} ila_ila2sir_next_t; + +typedef struct { + u32 ila_index; + ip6_address_t initial_dst; + u32 adj_index; +} ila_ila2sir_trace_t; + +static ila_entry_t ila_sir2ila_default_entry = { + .csum_mode = ILA_CSUM_MODE_NO_ACTION, + .type = ILA_TYPE_IID, + .dir = ILA_DIR_ILA2SIR, //Will pass the packet with no +}; + +/** + * @brief Dynamically registered DPO Type for ILA + */ +static dpo_type_t ila_dpo_type; + +/** + * @brief Dynamically registered FIB node type for ILA + */ +static fib_node_type_t ila_fib_node_type; + +u8 * +format_half_ip6_address (u8 * s, va_list * va) +{ + u64 v = clib_net_to_host_u64 (va_arg (*va, u64)); + + return format (s, "%04x:%04x:%04x:%04x", + v >> 48, (v >> 32) & 0xffff, (v >> 16) & 0xffff, v & 0xffff); + +} + +u8 * +format_ila_direction (u8 * s, va_list * args) +{ + ila_direction_t t = va_arg (*args, ila_direction_t); +#define _(i,n,st) \ + if (t == ILA_DIR_##i) \ + return format(s, st); + ila_foreach_direction +#undef _ + return format (s, "invalid_ila_direction"); +} + +static u8 * +format_csum_mode (u8 * s, va_list * va) +{ + ila_csum_mode_t csum_mode = va_arg (*va, ila_csum_mode_t); + char *txt; + + switch (csum_mode) + { +#define _(i,n,st) \ + case ILA_CSUM_MODE_##i: \ + txt = st; \ + break; + ila_csum_foreach_type +#undef _ + default: + txt = "invalid_ila_csum_mode"; + break; + } + return format (s, txt); +} + +u8 * +format_ila_type (u8 * s, va_list * args) +{ + ila_type_t t = va_arg (*args, ila_type_t); +#define _(i,n,st) \ + if (t == ILA_TYPE_##i) \ + return format(s, st); + ila_foreach_type +#undef _ + return format (s, "invalid_ila_type"); +} + +static u8 * +format_ila_entry (u8 * s, va_list * va) +{ + vnet_main_t *vnm = va_arg (*va, vnet_main_t *); + ila_entry_t *e = va_arg (*va, ila_entry_t *); + + if (!e) + { + return format (s, "%-15s%=40s%=40s%+16s%+18s%+11s", "Type", "SIR Address", + "ILA Address", "Checksum Mode", "Direction", "Next DPO"); + } + else if (vnm) + { + if (ip6_address_is_zero(&e->next_hop)) + { + return format (s, "%-15U%=40U%=40U%18U%11U%s", + format_ila_type, e->type, + format_ip6_address, &e->sir_address, + format_ip6_address, &e->ila_address, + format_csum_mode, e->csum_mode, + format_ila_direction, e->dir, + "n/a"); + } + else + { + return format (s, "%-15U%=40U%=40U%18U%11U%U", + format_ila_type, e->type, + format_ip6_address, &e->sir_address, + format_ip6_address, &e->ila_address, + format_csum_mode, e->csum_mode, + format_ila_direction, e->dir, + format_dpo_id, &e->ila_dpo, 0); + } + } + + return NULL; +} + +u8 * +format_ila_ila2sir_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 *); + ila_ila2sir_trace_t *t = va_arg (*args, ila_ila2sir_trace_t *); + return format (s, + "ILA -> SIR adj index: %d entry index: %d initial_dst: %U", + t->adj_index, t->ila_index, format_ip6_address, + &t->initial_dst); +} + +static uword +unformat_ila_direction (unformat_input_t * input, va_list * args) +{ + ila_direction_t *result = va_arg (*args, ila_direction_t *); +#define _(i,n,s) \ + if (unformat(input, s)) \ + { \ + *result = ILA_DIR_##i; \ + return 1;\ + } + + ila_foreach_direction +#undef _ + return 0; +} + +static uword +unformat_ila_type (unformat_input_t * input, va_list * args) +{ + ila_type_t *result = va_arg (*args, ila_type_t *); +#define _(i,n,s) \ + if (unformat(input, s)) \ + { \ + *result = ILA_TYPE_##i; \ + return 1;\ + } + + ila_foreach_type +#undef _ + return 0; +} + +static uword +unformat_ila_csum_mode (unformat_input_t * input, va_list * args) +{ + ila_csum_mode_t *result = va_arg (*args, ila_csum_mode_t *); + if (unformat (input, "none") || unformat (input, "no-action")) + { + *result = ILA_CSUM_MODE_NO_ACTION; + return 1; + } + if (unformat (input, "neutral-map")) + { + *result = ILA_CSUM_MODE_NEUTRAL_MAP; + return 1; + } + if (unformat (input, "adjust-transport")) + { + *result = ILA_CSUM_MODE_ADJUST_TRANSPORT; + return 1; + } + return 0; +} + +static uword +unformat_half_ip6_address (unformat_input_t * input, va_list * args) +{ + u64 *result = va_arg (*args, u64 *); + u32 a[4]; + + if (!unformat (input, "%x:%x:%x:%x", &a[0], &a[1], &a[2], &a[3])) + return 0; + + if (a[0] > 0xFFFF || a[1] > 0xFFFF || a[2] > 0xFFFF || a[3] > 0xFFFF) + return 0; + + *result = clib_host_to_net_u64 ((((u64) a[0]) << 48) | + (((u64) a[1]) << 32) | + (((u64) a[2]) << 16) | (((u64) a[3]))); + + return 1; +} + +static vlib_node_registration_t ila_ila2sir_node; + +static uword +ila_ila2sir (vlib_main_t * vm, + vlib_node_runtime_t * node, vlib_frame_t * frame) +{ + u32 n_left_from, *from, next_index, *to_next, n_left_to_next; + ila_main_t *ilm = &ila_main; + + from = vlib_frame_vector_args (frame); + n_left_from = frame->n_vectors; + next_index = node->cached_next_index; + + 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) + { + u32 pi0, pi1; + vlib_buffer_t *p0, *p1; + ila_entry_t *ie0, *ie1; + ip6_header_t *ip60, *ip61; + ip6_address_t *sir_address0, *sir_address1; + + { + 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, sizeof (ip6_header_t), LOAD); + CLIB_PREFETCH (p3->data, sizeof (ip6_header_t), LOAD); + } + + pi0 = to_next[0] = from[0]; + pi1 = to_next[1] = from[1]; + from += 2; + n_left_from -= 2; + to_next += 2; + n_left_to_next -= 2; + + p0 = vlib_get_buffer (vm, pi0); + p1 = vlib_get_buffer (vm, pi1); + ip60 = vlib_buffer_get_current (p0); + ip61 = vlib_buffer_get_current (p1); + sir_address0 = &ip60->dst_address; + sir_address1 = &ip61->dst_address; + ie0 = pool_elt_at_index (ilm->entries, + vnet_buffer (p0)->ip.adj_index[VLIB_TX]); + ie1 = pool_elt_at_index (ilm->entries, + vnet_buffer (p1)->ip.adj_index[VLIB_TX]); + + if (PREDICT_FALSE (p0->flags & VLIB_BUFFER_IS_TRACED)) + { + ila_ila2sir_trace_t *tr = + vlib_add_trace (vm, node, p0, sizeof (*tr)); + tr->ila_index = ie0 - ilm->entries; + tr->initial_dst = ip60->dst_address; + tr->adj_index = vnet_buffer (p0)->ip.adj_index[VLIB_TX]; + } + + if (PREDICT_FALSE (p1->flags & VLIB_BUFFER_IS_TRACED)) + { + ila_ila2sir_trace_t *tr = + vlib_add_trace (vm, node, p1, sizeof (*tr)); + tr->ila_index = ie1 - ilm->entries; + tr->initial_dst = ip61->dst_address; + tr->adj_index = vnet_buffer (p1)->ip.adj_index[VLIB_TX]; + } + + sir_address0 = (ie0->dir != ILA_DIR_SIR2ILA) ? &ie0->sir_address : sir_address0; + sir_address1 = (ie1->dir != ILA_DIR_SIR2ILA) ? &ie1->sir_address : sir_address1; + ip60->dst_address.as_u64[0] = sir_address0->as_u64[0]; + ip60->dst_address.as_u64[1] = sir_address0->as_u64[1]; + ip61->dst_address.as_u64[0] = sir_address1->as_u64[0]; + ip61->dst_address.as_u64[1] = sir_address1->as_u64[1]; + + vnet_buffer (p0)->ip.adj_index[VLIB_TX] = ie0->ila_dpo.dpoi_index; + vnet_buffer (p1)->ip.adj_index[VLIB_TX] = ie1->ila_dpo.dpoi_index; + + vlib_validate_buffer_enqueue_x2 (vm, node, next_index, to_next, + n_left_to_next, pi0, pi1, + ie0->ila_dpo.dpoi_next_node, + ie1->ila_dpo.dpoi_next_node); + } + + /* Single loop */ + while (n_left_from > 0 && n_left_to_next > 0) + { + u32 pi0; + vlib_buffer_t *p0; + ila_entry_t *ie0; + ip6_header_t *ip60; + ip6_address_t *sir_address0; + + pi0 = to_next[0] = from[0]; + from += 1; + n_left_from -= 1; + to_next += 1; + n_left_to_next -= 1; + + p0 = vlib_get_buffer (vm, pi0); + ip60 = vlib_buffer_get_current (p0); + sir_address0 = &ip60->dst_address; + ie0 = pool_elt_at_index (ilm->entries, + vnet_buffer (p0)->ip.adj_index[VLIB_TX]); + + if (PREDICT_FALSE (p0->flags & VLIB_BUFFER_IS_TRACED)) + { + ila_ila2sir_trace_t *tr = + vlib_add_trace (vm, node, p0, sizeof (*tr)); + tr->ila_index = ie0 ? (ie0 - ilm->entries) : ~0; + tr->initial_dst = ip60->dst_address; + tr->adj_index = vnet_buffer (p0)->ip.adj_index[VLIB_TX]; + } + + sir_address0 = (ie0->dir != ILA_DIR_SIR2ILA) ? &ie0->sir_address : sir_address0; + ip60->dst_address.as_u64[0] = sir_address0->as_u64[0]; + ip60->dst_address.as_u64[1] = sir_address0->as_u64[1]; + vnet_buffer (p0)->ip.adj_index[VLIB_TX] = ie0->ila_dpo.dpoi_index; + + vlib_validate_buffer_enqueue_x1 (vm, node, next_index, to_next, + n_left_to_next, pi0, + ie0->ila_dpo.dpoi_next_node); + } + vlib_put_next_frame (vm, node, next_index, n_left_to_next); + } + + return frame->n_vectors; +} + +/** *INDENT-OFF* */ +VLIB_REGISTER_NODE (ila_ila2sir_node, static) = +{ + .function = ila_ila2sir, + .name = "ila-to-sir", + .vector_size = sizeof (u32), + .format_trace = format_ila_ila2sir_trace, + .n_errors = ILA_N_ERROR, + .error_strings = ila_error_strings, + .n_next_nodes = ILA_ILA2SIR_N_NEXT, + .next_nodes = + { + [ILA_ILA2SIR_NEXT_DROP] = "error-drop" + }, +}; +/** *INDENT-ON* */ + +typedef enum +{ + ILA_SIR2ILA_NEXT_DROP, + ILA_SIR2ILA_N_NEXT, +} ila_sir2ila_next_t; + +typedef struct +{ + u32 ila_index; + ip6_address_t initial_dst; +} ila_sir2ila_trace_t; + +u8 * +format_ila_sir2ila_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 *); + ila_sir2ila_trace_t *t = va_arg (*args, ila_sir2ila_trace_t *); + + return format (s, "SIR -> ILA entry index: %d initial_dst: %U", + t->ila_index, format_ip6_address, &t->initial_dst); +} + +static vlib_node_registration_t ila_sir2ila_node; + +static uword +ila_sir2ila (vlib_main_t * vm, + vlib_node_runtime_t * node, vlib_frame_t * frame) +{ + u32 n_left_from, *from, next_index, *to_next, n_left_to_next; + ila_main_t *ilm = &ila_main; + + from = vlib_frame_vector_args (frame); + n_left_from = frame->n_vectors; + next_index = node->cached_next_index; + + 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) + { + u32 pi0, pi1; + vlib_buffer_t *p0, *p1; + ip6_header_t *ip60, *ip61; + u32 next0 = ILA_SIR2ILA_NEXT_DROP; + u32 next1 = ILA_SIR2ILA_NEXT_DROP; + BVT (clib_bihash_kv) kv0, value0; + BVT (clib_bihash_kv) kv1, value1; + ila_entry_t *ie0 = &ila_sir2ila_default_entry; + ila_entry_t *ie1 = &ila_sir2ila_default_entry; + ip6_address_t *ila_address0, *ila_address1; + + { + 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, sizeof (ip6_header_t), LOAD); + CLIB_PREFETCH (p3->data, sizeof (ip6_header_t), LOAD); + } + + pi0 = to_next[0] = from[0]; + pi1 = to_next[1] = from[1]; + from += 2; + n_left_from -= 2; + to_next += 2; + n_left_to_next -= 2; + + p0 = vlib_get_buffer (vm, pi0); + p1 = vlib_get_buffer (vm, pi1); + ip60 = vlib_buffer_get_current (p0); + ip61 = vlib_buffer_get_current (p1); + ila_address0 = &ip60->dst_address; + ila_address1 = &ip61->dst_address; + kv0.key[0] = ip60->dst_address.as_u64[0]; + kv0.key[1] = ip60->dst_address.as_u64[1]; + kv0.key[2] = 0; + kv1.key[0] = ip61->dst_address.as_u64[0]; + kv1.key[1] = ip61->dst_address.as_u64[1]; + kv1.key[2] = 0; + + if (PREDICT_TRUE((BV (clib_bihash_search) + (&ilm->id_to_entry_table, &kv0, &value0)) == 0)) { + ie0 = &ilm->entries[value0.value]; + ila_address0 = (ie0->dir != ILA_DIR_ILA2SIR) ? &ie0->ila_address : ila_address0; + } + + if ((BV (clib_bihash_search) + (&ilm->id_to_entry_table, &kv1, &value1)) == 0) { + ie1 = &ilm->entries[value1.value]; + ila_address1 = (ie1->dir != ILA_DIR_ILA2SIR) ? &ie1->ila_address : ila_address1; + } + + if (PREDICT_FALSE (p0->flags & VLIB_BUFFER_IS_TRACED)) + { + ila_sir2ila_trace_t *tr = + vlib_add_trace (vm, node, p0, sizeof (*tr)); + tr->ila_index = + (ie0 != &ila_sir2ila_default_entry) ? (ie0 - ilm->entries) : ~0; + tr->initial_dst = ip60->dst_address; + } + + if (PREDICT_FALSE (p1->flags & VLIB_BUFFER_IS_TRACED)) + { + ila_sir2ila_trace_t *tr = + vlib_add_trace (vm, node, p1, sizeof (*tr)); + tr->ila_index = + (ie1 != &ila_sir2ila_default_entry) ? (ie1 - ilm->entries) : ~0; + tr->initial_dst = ip61->dst_address; + } + + ip60->dst_address.as_u64[0] = ila_address0->as_u64[0]; + ip60->dst_address.as_u64[1] = ila_address0->as_u64[1]; + ip61->dst_address.as_u64[0] = ila_address1->as_u64[0]; + ip61->dst_address.as_u64[1] = ila_address1->as_u64[1]; + + vnet_feature_next (vnet_buffer (p0)->sw_if_index[VLIB_RX], &next0, p0); + vnet_feature_next (vnet_buffer (p1)->sw_if_index[VLIB_RX], &next1, p1); + + vlib_validate_buffer_enqueue_x2 (vm, node, next_index, to_next, + n_left_to_next, pi0, pi1, next0, + next1); + } + + /* Single loop */ + while (n_left_from > 0 && n_left_to_next > 0) + { + u32 pi0; + vlib_buffer_t *p0; + ip6_header_t *ip60; + u32 next0 = ILA_SIR2ILA_NEXT_DROP; + BVT (clib_bihash_kv) kv0, value0; + ila_entry_t *ie0 = &ila_sir2ila_default_entry; + ip6_address_t *ila_address0; + + pi0 = to_next[0] = from[0]; + from += 1; + n_left_from -= 1; + to_next += 1; + n_left_to_next -= 1; + + p0 = vlib_get_buffer (vm, pi0); + ip60 = vlib_buffer_get_current (p0); + ila_address0 = &ip60->dst_address; + + kv0.key[0] = ip60->dst_address.as_u64[0]; + kv0.key[1] = ip60->dst_address.as_u64[1]; + kv0.key[2] = 0; + + if (PREDICT_TRUE((BV (clib_bihash_search) + (&ilm->id_to_entry_table, &kv0, &value0)) == 0)) { + ie0 = &ilm->entries[value0.value]; + ila_address0 = (ie0->dir != ILA_DIR_ILA2SIR) ? &ie0->ila_address : ila_address0; + } + + if (PREDICT_FALSE (p0->flags & VLIB_BUFFER_IS_TRACED)) + { + ila_sir2ila_trace_t *tr = + vlib_add_trace (vm, node, p0, sizeof (*tr)); + tr->ila_index = + (ie0 != &ila_sir2ila_default_entry) ? (ie0 - ilm->entries) : ~0; + tr->initial_dst = ip60->dst_address; + } + + //This operation should do everything for any type (except vnid4 obviously) + ip60->dst_address.as_u64[0] = ila_address0->as_u64[0]; + ip60->dst_address.as_u64[1] = ila_address0->as_u64[1]; + + vnet_feature_next (vnet_buffer (p0)->sw_if_index[VLIB_RX], &next0, p0); + + vlib_validate_buffer_enqueue_x1 (vm, node, next_index, to_next, + n_left_to_next, pi0, next0); + } + vlib_put_next_frame (vm, node, next_index, n_left_to_next); + } + + return frame->n_vectors; +} + +/** *INDENT-OFF* */ +VLIB_REGISTER_NODE (ila_sir2ila_node, static) = +{ + .function = ila_sir2ila,.name = "sir-to-ila", + .vector_size = sizeof (u32), + .format_trace = format_ila_sir2ila_trace, + .n_errors = ILA_N_ERROR, + .error_strings = ila_error_strings, + .n_next_nodes = ILA_SIR2ILA_N_NEXT, + .next_nodes = + { + [ILA_SIR2ILA_NEXT_DROP] = "error-drop" + }, +}; +/** *INDENT-ON* */ + +/** *INDENT-OFF* */ +VNET_FEATURE_INIT (ila_sir2ila, static) = +{ + .arc_name = "ip6-unicast", + .node_name = "sir-to-ila", + .runs_before = VNET_FEATURES ("ip6-lookup"), +}; +/** *INDENT-ON* */ + +static void +ila_entry_stack (ila_entry_t *ie) +{ + /* + * restack on the next-hop's FIB entry + */ + dpo_stack(ila_dpo_type, + DPO_PROTO_IP6, + &ie->ila_dpo, + fib_entry_contribute_ip_forwarding( + ie->next_hop_fib_entry_index)); +} + +int +ila_add_del_entry (ila_add_del_entry_args_t * args) +{ + ila_main_t *ilm = &ila_main; + BVT (clib_bihash_kv) kv, value; + + //Sanity check + if (args->type == ILA_TYPE_IID || args->type == ILA_TYPE_LUID) + { + if ((args->sir_address.as_u8[8] >> 5) != args->type) + { + clib_warning ("Incorrect SIR address (ILA type mismatch %d %d)", + args->sir_address.as_u8[8] >> 1, args->type); + return -1; + } + if (args->sir_address.as_u8[8] & 0x10) + { + clib_warning ("Checksum bit should not be set in SIR address"); + return -1; + } + } + else if (args->type == ILA_TYPE_VNIDM) + { + if (args->sir_address.as_u8[0] != 0xff || + (args->sir_address.as_u8[1] & 0xf0) != 0xf0) + { + clib_warning ("SIR multicast address must start with fff"); + return -1; + } + if (args->sir_address.as_u16[1] || args->sir_address.as_u16[2] || + args->sir_address.as_u16[3] || args->sir_address.as_u16[4] || + args->sir_address.as_u16[5] || (args->sir_address.as_u8[12] & 0xf0)) + { + clib_warning ("SIR multicast address must start with fff"); + return -1; + } + } + + if (!args->is_del) + { + ila_entry_t *e; + pool_get (ilm->entries, e); + e->type = args->type; + e->sir_address = args->sir_address; + e->next_hop = args->next_hop_address; + e->csum_mode = args->csum_mode; + e->dir = args->dir; + + //Construct ILA address + switch (e->type) + { + case ILA_TYPE_IID: + e->ila_address = e->sir_address; + break; + case ILA_TYPE_LUID: + e->ila_address.as_u64[0] = args->locator; + e->ila_address.as_u64[1] = args->sir_address.as_u64[1]; + break; + case ILA_TYPE_VNID6: + e->ila_address.as_u64[0] = args->locator; + e->ila_address.as_u8[8] = (ILA_TYPE_VNID6 << 1); + e->ila_address.as_u32[2] |= args->vnid; + e->ila_address.as_u32[3] = args->sir_address.as_u32[3]; + break; + case ILA_TYPE_VNIDM: + e->ila_address.as_u64[0] = args->locator; + e->ila_address.as_u8[8] = (ILA_TYPE_VNIDM << 1); + e->ila_address.as_u32[2] |= args->vnid; + e->ila_address.as_u32[3] = args->sir_address.as_u32[3]; + e->ila_address.as_u8[12] |= args->sir_address.as_u8[2] << 4; + break; + case ILA_TYPE_VNID4: + clib_warning ("ILA type '%U' is not supported", format_ila_type, + e->type); + return -1; + } + + //Modify ILA checksum if necessary + if (e->csum_mode == ILA_CSUM_MODE_NEUTRAL_MAP) + { + ip_csum_t csum = e->ila_address.as_u16[7]; + int i; + for (i = 0; i < 4; i++) + { + csum = ip_csum_sub_even (csum, e->sir_address.as_u32[i]); + csum = ip_csum_add_even (csum, e->ila_address.as_u32[i]); + } + csum = ip_csum_add_even (csum, clib_host_to_net_u16 (0x1000)); + e->ila_address.as_u16[7] = ip_csum_fold (csum); + e->ila_address.as_u8[8] |= 0x10; + } + + //Create entry with the sir address + kv.key[0] = e->sir_address.as_u64[0]; + kv.key[1] = e->sir_address.as_u64[1]; + kv.key[2] = 0; + kv.value = e - ilm->entries; + BV (clib_bihash_add_del) (&ilm->id_to_entry_table, &kv, + 1 /* is_add */ ); + + if (!ip6_address_is_zero(&e->next_hop)) + { + /* + * become a child of the FIB netry for the next-hop + * so we are informed when its forwarding changes + */ + fib_prefix_t next_hop = { + .fp_addr = { + .ip6 = e->next_hop, + }, + .fp_len = 128, + .fp_proto = FIB_PROTOCOL_IP6, + }; + + e->next_hop_fib_entry_index = + fib_table_entry_special_add(0, + &next_hop, + FIB_SOURCE_RR, + FIB_ENTRY_FLAG_NONE, + ADJ_INDEX_INVALID); + e->next_hop_child_index = + fib_entry_child_add(e->next_hop_fib_entry_index, + ila_fib_node_type, + e - ilm->entries); + + /* + * Create a route that results in the ILA entry + */ + dpo_id_t dpo = DPO_INVALID; + fib_prefix_t pfx = { + .fp_addr = { + .ip6 = e->ila_address, + }, + .fp_len = 128, + .fp_proto = FIB_PROTOCOL_IP6, + }; + + dpo_set(&dpo, ila_dpo_type, DPO_PROTO_IP6, e - ilm->entries); + + fib_table_entry_special_dpo_add(0, + &pfx, + FIB_SOURCE_PLUGIN_HI, + FIB_ENTRY_FLAG_EXCLUSIVE, + &dpo); + dpo_reset(&dpo); + + /* + * finally stack the ILA entry so it will forward to the next-hop + */ + ila_entry_stack (e); + } + } + else + { + ila_entry_t *e; + kv.key[0] = args->sir_address.as_u64[0]; + kv.key[1] = args->sir_address.as_u64[1]; + kv.key[2] = 0; + + if ((BV (clib_bihash_search) (&ilm->id_to_entry_table, &kv, &value) < + 0)) + { + return -1; + } + + e = &ilm->entries[value.value]; + + if (!ip6_address_is_zero(&e->next_hop)) + { + fib_prefix_t pfx = { + .fp_addr = { + .ip6 = e->ila_address, + }, + .fp_len = 128, + .fp_proto = FIB_PROTOCOL_IP6, + }; + + fib_table_entry_special_remove(0, &pfx, FIB_SOURCE_PLUGIN_HI); + /* + * remove this ILA entry as child of the FIB netry for the next-hop + */ + fib_entry_child_remove(e->next_hop_fib_entry_index, + e->next_hop_child_index); + fib_table_entry_delete_index(e->next_hop_fib_entry_index, + FIB_SOURCE_RR); + e->next_hop_fib_entry_index = FIB_NODE_INDEX_INVALID; + } + dpo_reset (&e->ila_dpo); + + BV (clib_bihash_add_del) (&ilm->id_to_entry_table, &kv, + 0 /* is_add */ ); + pool_put (ilm->entries, e); + } + return 0; +} + +int +ila_interface (u32 sw_if_index, u8 disable) +{ + vnet_feature_enable_disable ("ip4-unicast", "sir-to-ila", sw_if_index, + !disable, 0, 0); + return 0; +} + +clib_error_t * +vlib_plugin_register (vlib_main_t * vm, vnet_plugin_handoff_t * h, + int from_early_init) +{ + clib_error_t *error = 0; + + return error; +} + +u8 *format_ila_dpo (u8 * s, va_list * va) +{ + index_t index = va_arg (*va, index_t); + CLIB_UNUSED(u32 indent) = va_arg (*va, u32); + ila_main_t *ilm = &ila_main; + ila_entry_t *ie = pool_elt_at_index (ilm->entries, index); + return format(s, "ILA: idx:%d sir:%U", + index, + format_ip6_address, &ie->sir_address); +} + +/** + * @brief no-op lock function. + * The lifetime of the ILA entry is managed by the control plane + */ +static void +ila_dpo_lock (dpo_id_t *dpo) +{ +} + +/** + * @brief no-op unlock function. + * The lifetime of the ILA entry is managed by the control plane + */ +static void +ila_dpo_unlock (dpo_id_t *dpo) +{ +} + +const static dpo_vft_t ila_vft = { + .dv_lock = ila_dpo_lock, + .dv_unlock = ila_dpo_unlock, + .dv_format = format_ila_dpo, +}; +const static char* const ila_ip6_nodes[] = +{ + "ila-to-sir", + NULL, +}; +const static char* const * const ila_nodes[DPO_PROTO_NUM] = +{ + [DPO_PROTO_IP6] = ila_ip6_nodes, +}; + +static fib_node_t * +ila_fib_node_get_node (fib_node_index_t index) +{ + ila_main_t *ilm = &ila_main; + ila_entry_t *ie = pool_elt_at_index (ilm->entries, index); + + return (&ie->ila_fib_node); +} + +/** + * @brief no-op unlock function. + * The lifetime of the ILA entry is managed by the control plane + */ +static void +ila_fib_node_last_lock_gone (fib_node_t *node) +{ +} + +static ila_entry_t * +ila_entry_from_fib_node (fib_node_t *node) +{ + return ((ila_entry_t*)(((char*)node) - + STRUCT_OFFSET_OF(ila_entry_t, ila_fib_node))); +} + +/** + * @brief + * Callback function invoked when the forwarding changes for the ILA next-hop + */ +static fib_node_back_walk_rc_t +ila_fib_node_back_walk_notify (fib_node_t *node, + fib_node_back_walk_ctx_t *ctx) +{ + ila_entry_stack(ila_entry_from_fib_node(node)); + + return (FIB_NODE_BACK_WALK_CONTINUE); +} + +/* + * ILA's FIB graph node virtual function table + */ +static const fib_node_vft_t ila_fib_node_vft = { + .fnv_get = ila_fib_node_get_node, + .fnv_last_lock = ila_fib_node_last_lock_gone, + .fnv_back_walk = ila_fib_node_back_walk_notify, +}; + +clib_error_t * +ila_init (vlib_main_t * vm) +{ + ila_main_t *ilm = &ila_main; + ilm->entries = NULL; + + ilm->lookup_table_nbuckets = ILA_TABLE_DEFAULT_HASH_NUM_BUCKETS; + ilm->lookup_table_nbuckets = 1 << max_log2 (ilm->lookup_table_nbuckets); + ilm->lookup_table_size = ILA_TABLE_DEFAULT_HASH_MEMORY_SIZE; + + BV (clib_bihash_init) (&ilm->id_to_entry_table, + "ila id to entry index table", + ilm->lookup_table_nbuckets, ilm->lookup_table_size); + + ila_dpo_type = dpo_register_new_type(&ila_vft, ila_nodes); + ila_fib_node_type = fib_node_register_new_type(&ila_fib_node_vft); + + return NULL; +} + +VLIB_INIT_FUNCTION (ila_init); + +static clib_error_t * +ila_entry_command_fn (vlib_main_t * vm, + unformat_input_t * input, vlib_cli_command_t * cmd) +{ + unformat_input_t _line_input, *line_input = &_line_input; + ila_add_del_entry_args_t args = { 0 }; + u8 next_hop_set = 0; + int ret; + + args.type = ILA_TYPE_IID; + args.csum_mode = ILA_CSUM_MODE_NO_ACTION; + args.local_adj_index = ~0; + args.dir = ILA_DIR_BIDIR; + + if (!unformat_user (input, unformat_line_input, line_input)) + return 0; + + while (unformat_check_input (line_input) != UNFORMAT_END_OF_INPUT) + { + if (unformat (line_input, "type %U", unformat_ila_type, &args.type)) + ; + else if (unformat + (line_input, "sir-address %U", unformat_ip6_address, + &args.sir_address)) + ; + else if (unformat + (line_input, "locator %U", unformat_half_ip6_address, + &args.locator)) + ; + else if (unformat + (line_input, "csum-mode %U", unformat_ila_csum_mode, + &args.csum_mode)) + ; + else if (unformat (line_input, "vnid %x", &args.vnid)) + ; + else if (unformat + (line_input, "next-hop %U", unformat_ip6_address, + &args.next_hop_address)) + ; + else if (unformat + (line_input, "direction %U", unformat_ila_direction, &args.dir)) + next_hop_set = 1; + else if (unformat (line_input, "del")) + args.is_del = 1; + else + return clib_error_return (0, "parse error: '%U'", + format_unformat_error, line_input); + } + + unformat_free (line_input); + + if (!next_hop_set) + return clib_error_return (0, "Specified a next hop"); + + if ((ret = ila_add_del_entry (&args))) + return clib_error_return (0, "ila_add_del_entry returned error %d", ret); + + return NULL; +} + +VLIB_CLI_COMMAND (ila_entry_command, static) = +{ + .path = "ila entry", + .short_help = "ila entry [type <type>] [sir-address <address>] [locator <locator>] [vnid <hex-vnid>]" + " [adj-index <adj-index>] [next-hop <next-hop>] [direction (bidir|sir2ila|ila2sir)]" + " [csum-mode (no-action|neutral-map|transport-adjust)] [del]", + .function = ila_entry_command_fn, +}; + +static clib_error_t * +ila_interface_command_fn (vlib_main_t * vm, + unformat_input_t * input, vlib_cli_command_t * cmd) +{ + vnet_main_t *vnm = vnet_get_main (); + u32 sw_if_index = ~0; + u8 disable = 0; + + if (!unformat (input, "%U", unformat_vnet_sw_interface, vnm, &sw_if_index)) + { + return clib_error_return (0, "Invalid interface name"); + } + + if (unformat (input, "disable")) + { + disable = 1; + } + + int ret; + if ((ret = ila_interface (sw_if_index, disable))) + return clib_error_return (0, "ila_interface returned error %d", ret); + + return NULL; +} + +VLIB_CLI_COMMAND (ila_interface_command, static) = +{ + .path = "ila interface", + .short_help = "ila interface <interface-name> [disable]", + .function = ila_interface_command_fn, +}; + +static clib_error_t * +ila_show_entries_command_fn (vlib_main_t * vm, + unformat_input_t * input, + vlib_cli_command_t * cmd) +{ + vnet_main_t *vnm = vnet_get_main (); + ila_main_t *ilm = &ila_main; + ila_entry_t *e; + + vlib_cli_output (vm, " %U\n", format_ila_entry, vnm, NULL); + pool_foreach (e, ilm->entries, + ({ + vlib_cli_output (vm, " %U\n", format_ila_entry, vnm, e); + })); + + return NULL; +} + +VLIB_CLI_COMMAND (ila_show_entries_command, static) = +{ + .path = "show ila entries", + .short_help = "show ila entries", + .function = ila_show_entries_command_fn, +}; diff --git a/src/plugins/ila/ila.h b/src/plugins/ila/ila.h new file mode 100644 index 00000000000..26620983823 --- /dev/null +++ b/src/plugins/ila/ila.h @@ -0,0 +1,116 @@ +/* + * 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: + * + * 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. + */ + +#ifndef ILA_H +#define ILA_H + +#include <vnet/vnet.h> +#include <vnet/ip/ip.h> +#include <vnet/fib/fib_node.h> + +#include <vppinfra/bihash_24_8.h> +#include <vppinfra/bihash_template.h> + +#define ila_foreach_type \ + _(IID, 0, "iid") \ + _(LUID, 1, "luid") \ + _(VNID4, 2, "vnid-ip4") \ + _(VNID6, 3, "vnid-ip6") \ + _(VNIDM, 4, "vnid-multicast") + +typedef enum { +#define _(i,n,s) ILA_TYPE_##i = n, + ila_foreach_type +#undef _ +} ila_type_t; + +#define ila_csum_foreach_type \ +_(NO_ACTION, 0, "no-action") \ +_(NEUTRAL_MAP, 1, "neutral-map") \ +_(ADJUST_TRANSPORT, 2, "adjust-transport") + +typedef enum { +#define _(i,n,s) ILA_CSUM_MODE_##i = n, + ila_csum_foreach_type +#undef _ + ILA_CSUM_N_TYPES +} ila_csum_mode_t; + +#define ila_foreach_direction \ +_(BIDIR, 0, "bidir") \ +_(SIR2ILA, 1, "sir2ila") \ +_(ILA2SIR, 2, "ila2sir") + +typedef enum { +#define _(i,n,s) ILA_DIR_##i = n, + ila_foreach_direction +#undef _ +} ila_direction_t; + +typedef struct { + /** + * Fib Node base class + */ + fib_node_t ila_fib_node; + ila_type_t type; + ip6_address_t sir_address; + ip6_address_t ila_address; + ip6_address_t next_hop; + ila_csum_mode_t csum_mode; + ila_direction_t dir; + + /** + * The FIB entry index for the next-hop + */ + fib_node_index_t next_hop_fib_entry_index; + + /** + * The child index on the FIB entry + */ + u32 next_hop_child_index; + + /** + * The next DPO in the grpah to follow + */ + dpo_id_t ila_dpo; +} ila_entry_t; + +typedef struct { + ila_entry_t *entries; //Pool of ILA entries + + u64 lookup_table_nbuckets; + u64 lookup_table_size; + clib_bihash_24_8_t id_to_entry_table; + + u32 ip6_lookup_next_index; +} ila_main_t; + + +typedef struct { + ila_type_t type; + ip6_address_t sir_address; + ip6_address_t next_hop_address; + u64 locator; + u32 vnid; + u32 local_adj_index; + ila_csum_mode_t csum_mode; + ila_direction_t dir; + u8 is_del; +} ila_add_del_entry_args_t; + +int ila_add_del_entry (ila_add_del_entry_args_t * args); +int ila_interface (u32 sw_if_index, u8 disable); + +#endif //ILA_H diff --git a/src/plugins/sixrd.am b/src/plugins/sixrd.am new file mode 100644 index 00000000000..0de4508831e --- /dev/null +++ b/src/plugins/sixrd.am @@ -0,0 +1,26 @@ +# Copyright (c) 2015 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. + +libsixrd_plugin_la_SOURCES = \ + sixrd/sixrd.c \ + sixrd/sixrd_dpo.c \ + sixrd/ip4_sixrd.c \ + sixrd/ip6_sixrd.c + +noinst_HEADERS += \ + sixrd/sixrd.h \ + sixrd/sixrd_dpo.h + +vppplugins_LTLIBRARIES += libsixrd_plugin.la + +# vi:syntax=automake diff --git a/src/plugins/sixrd/ip4_sixrd.c b/src/plugins/sixrd/ip4_sixrd.c new file mode 100644 index 00000000000..2fb8015d994 --- /dev/null +++ b/src/plugins/sixrd/ip4_sixrd.c @@ -0,0 +1,127 @@ +/*--------------------------------------------------------------------------- + * Copyright (c) 2009-2014 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 "sixrd.h" + +static vlib_node_registration_t ip4_sixrd_node; + +typedef enum { + IP4_SIXRD_NEXT_IP6_LOOKUP, + IP4_SIXRD_NEXT_DROP, + IP4_SIXRD_N_NEXT, +} ip4_sixrd_next_t; + +/* + * ip4_sixrd_sec_check + */ +static_always_inline void +ip4_sixrd_sec_check (sixrd_domain_t *d, ip4_address_t sa4, ip6_address_t sa6, u8 *error) +{ + u32 a = sixrd_get_addr(d, sa6.as_u64[0]); + clib_warning("Security check: %U %U", format_ip4_address, &a, format_ip4_address, &sa4); + if (PREDICT_FALSE(sixrd_get_addr(d, sa6.as_u64[0]) != sa4.as_u32)) + *error = SIXRD_ERROR_SEC_CHECK; +} + +/* + * ip4_sixrd + */ +static uword +ip4_sixrd (vlib_main_t *vm, + vlib_node_runtime_t *node, + vlib_frame_t *frame) +{ + u32 n_left_from, *from, next_index, *to_next, n_left_to_next; + vlib_node_runtime_t *error_node = vlib_node_get_runtime(vm, ip4_sixrd_node.index); + u32 decap = 0; + + from = vlib_frame_vector_args(frame); + n_left_from = frame->n_vectors; + next_index = node->cached_next_index; + while (n_left_from > 0) { + vlib_get_next_frame(vm, node, next_index, to_next, n_left_to_next); + + /* Single loop */ + while (n_left_from > 0 && n_left_to_next > 0) { + u32 pi0; + vlib_buffer_t *p0; + u8 error0 = SIXRD_ERROR_NONE; + sixrd_domain_t *d0 = 0; + ip4_header_t *ip40; + ip6_header_t *ip60; + u32 sixrd_domain_index0 = ~0; + u32 next0; + + pi0 = to_next[0] = from[0]; + from += 1; + n_left_from -= 1; + to_next +=1; + n_left_to_next -= 1; + + p0 = vlib_get_buffer(vm, pi0); + ip40 = vlib_buffer_get_current(p0); + + /* Throw away anything that isn't IP in IP. */ + if (PREDICT_TRUE(ip40->protocol == IP_PROTOCOL_IPV6 && clib_net_to_host_u16(ip40->length) >= 60)) { + vlib_buffer_advance(p0, sizeof(ip4_header_t)); + ip60 = vlib_buffer_get_current(p0); + d0 = ip4_sixrd_get_domain(vnet_buffer(p0)->ip.adj_index[VLIB_TX], (ip6_address_t *)&ip60->src_address, + &sixrd_domain_index0, &error0); + } else { + error0 = SIXRD_ERROR_BAD_PROTOCOL; + } + if (d0) { + /* SIXRD inbound security check */ + ip4_sixrd_sec_check(d0, ip40->src_address, ip60->src_address, &error0); + } + + next0 = error0 == SIXRD_ERROR_NONE ? IP4_SIXRD_NEXT_IP6_LOOKUP : IP4_SIXRD_NEXT_DROP; + + if (PREDICT_FALSE(p0->flags & VLIB_BUFFER_IS_TRACED)) { + sixrd_trace_t *tr = vlib_add_trace(vm, node, p0, sizeof(*tr)); + tr->sixrd_domain_index = sixrd_domain_index0; + } + + p0->error = error_node->errors[error0]; + if (PREDICT_TRUE(error0 == SIXRD_ERROR_NONE)) decap++; + vlib_validate_buffer_enqueue_x1(vm, node, next_index, to_next, n_left_to_next, pi0, next0); + + } + vlib_put_next_frame(vm, node, next_index, n_left_to_next); + } + vlib_node_increment_counter(vm, ip4_sixrd_node.index, SIXRD_ERROR_DECAPSULATED, decap); + + return frame->n_vectors; +} + +static char *sixrd_error_strings[] = { +#define _(sym,string) string, + foreach_sixrd_error +#undef _ +}; + +VLIB_REGISTER_NODE(ip4_sixrd_node,static) = { + .function = ip4_sixrd, + .name = "ip4-sixrd", + .vector_size = sizeof(u32), + .format_trace = format_sixrd_trace, + .n_errors = SIXRD_N_ERROR, + .error_strings = sixrd_error_strings, + .n_next_nodes = IP4_SIXRD_N_NEXT, + .next_nodes = { + [IP4_SIXRD_NEXT_IP6_LOOKUP] = "ip6-lookup", + [IP4_SIXRD_NEXT_DROP] = "error-drop", + }, +}; diff --git a/src/plugins/sixrd/ip6_sixrd.c b/src/plugins/sixrd/ip6_sixrd.c new file mode 100644 index 00000000000..36f3fab320b --- /dev/null +++ b/src/plugins/sixrd/ip6_sixrd.c @@ -0,0 +1,129 @@ +/*--------------------------------------------------------------------------- + * Copyright (c) 2009-2014 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. + *--------------------------------------------------------------------------- + */ +/* + * Defines used for testing various optimisation schemes + */ +#define SIXRD_ENCAP_DUAL 0 + +#include "sixrd.h" + +static vlib_node_registration_t ip6_sixrd_node; + +typedef enum { + IP6_SIXRD_NEXT_IP4_LOOKUP, + IP6_SIXRD_NEXT_DROP, + IP6_SIXRD_N_NEXT, +} ip6_sixrd_next_t; + +/* + * ip6_sixrd + */ +static uword +ip6_sixrd (vlib_main_t *vm, + vlib_node_runtime_t *node, + vlib_frame_t *frame) +{ + u32 n_left_from, *from, next_index, *to_next, n_left_to_next; + vlib_node_runtime_t *error_node = vlib_node_get_runtime(vm, ip6_sixrd_node.index); + u32 encap = 0; + from = vlib_frame_vector_args(frame); + n_left_from = frame->n_vectors; + next_index = node->cached_next_index; + + while (n_left_from > 0) { + vlib_get_next_frame(vm, node, next_index, to_next, n_left_to_next); + + while (n_left_from > 0 && n_left_to_next > 0) { + u32 pi0; + vlib_buffer_t *p0; + sixrd_domain_t *d0; + u8 error0 = SIXRD_ERROR_NONE; + ip6_header_t *ip60; + ip4_header_t *ip4h0; + u32 next0 = IP6_SIXRD_NEXT_IP4_LOOKUP; + u32 sixrd_domain_index0 = ~0; + + pi0 = to_next[0] = from[0]; + from += 1; + n_left_from -= 1; + to_next +=1; + n_left_to_next -= 1; + + p0 = vlib_get_buffer(vm, pi0); + ip60 = vlib_buffer_get_current(p0); + // p0->current_length = clib_net_to_host_u16(ip40->length); + d0 = ip6_sixrd_get_domain(vnet_buffer(p0)->ip.adj_index[VLIB_TX], &sixrd_domain_index0); + ASSERT(d0); + + /* SIXRD calc */ + u64 dal60 = clib_net_to_host_u64(ip60->dst_address.as_u64[0]); + u32 da40 = sixrd_get_addr(d0, dal60); + u16 len = clib_net_to_host_u16(ip60->payload_length) + 60; + if (da40 == 0) error0 = SIXRD_ERROR_UNKNOWN; + + /* construct ipv4 header */ + vlib_buffer_advance(p0, - (sizeof(ip4_header_t))); + ip4h0 = vlib_buffer_get_current(p0); + vnet_buffer(p0)->sw_if_index[VLIB_TX] = (u32)~0; + ip4h0->ip_version_and_header_length = 0x45; + ip4h0->tos = 0; + ip4h0->length = clib_host_to_net_u16(len); + ip4h0->fragment_id = 0; + ip4h0->flags_and_fragment_offset = 0; + ip4h0->ttl = 0x40; + ip4h0->protocol = IP_PROTOCOL_IPV6; + ip4h0->src_address = d0->ip4_src; + ip4h0->dst_address.as_u32 = clib_host_to_net_u32(da40); + ip4h0->checksum = ip4_header_checksum(ip4h0); + + next0 = error0 == SIXRD_ERROR_NONE ? IP6_SIXRD_NEXT_IP4_LOOKUP : IP6_SIXRD_NEXT_DROP; + + if (PREDICT_FALSE(p0->flags & VLIB_BUFFER_IS_TRACED)) { + sixrd_trace_t *tr = vlib_add_trace(vm, node, p0, sizeof(*tr)); + tr->sixrd_domain_index = sixrd_domain_index0; + } + + p0->error = error_node->errors[error0]; + if (PREDICT_TRUE(error0 == SIXRD_ERROR_NONE)) encap++; + + vlib_validate_buffer_enqueue_x1(vm, node, next_index, to_next, n_left_to_next, pi0, next0); + } + vlib_put_next_frame(vm, node, next_index, n_left_to_next); + } + vlib_node_increment_counter(vm, ip6_sixrd_node.index, SIXRD_ERROR_ENCAPSULATED, encap); + + return frame->n_vectors; +} + +static char *sixrd_error_strings[] = { +#define _(sym,string) string, + foreach_sixrd_error +#undef _ +}; + +VLIB_REGISTER_NODE(ip6_sixrd_node,static) = { + .function = ip6_sixrd, + .name = "ip6-sixrd", + .vector_size = sizeof(u32), + .format_trace = format_sixrd_trace, + .n_errors = SIXRD_N_ERROR, + .error_strings = sixrd_error_strings, + .n_next_nodes = IP6_SIXRD_N_NEXT, + .next_nodes = { + [IP6_SIXRD_NEXT_IP4_LOOKUP] = "ip4-lookup", + [IP6_SIXRD_NEXT_DROP] = "error-drop", + }, +}; diff --git a/src/plugins/sixrd/sixrd.c b/src/plugins/sixrd/sixrd.c new file mode 100644 index 00000000000..66e631a2b6a --- /dev/null +++ b/src/plugins/sixrd/sixrd.c @@ -0,0 +1,369 @@ +/* + * Copyright (c) 2015 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 "sixrd.h" +#include <vnet/plugin/plugin.h> + +#include <vnet/fib/fib_table.h> +#include <vnet/fib/ip6_fib.h> +#include <vnet/adj/adj.h> + +/* + * This code supports the following sixrd modes: + * + * 32 EA bits (Complete IPv4 address is embedded): + * ea_bits_len = 32 + * IPv4 suffix is embedded: + * ea_bits_len = < 32 + * No embedded address bits (1:1 mode): + * ea_bits_len = 0 + */ + +int +sixrd_create_domain (ip6_address_t *ip6_prefix, + u8 ip6_prefix_len, + ip4_address_t *ip4_prefix, + u8 ip4_prefix_len, + ip4_address_t *ip4_src, + u32 *sixrd_domain_index, + u16 mtu) +{ + dpo_id_t dpo_v6 = DPO_INVALID, dpo_v4 = DPO_INVALID; + sixrd_main_t *mm = &sixrd_main; + fib_node_index_t fei; + sixrd_domain_t *d; + + /* Get domain index */ + pool_get_aligned(mm->domains, d, CLIB_CACHE_LINE_BYTES); + memset(d, 0, sizeof (*d)); + *sixrd_domain_index = d - mm->domains; + + /* Init domain struct */ + d->ip4_prefix.as_u32 = ip4_prefix->as_u32; + d->ip4_prefix_len = ip4_prefix_len; + d->ip6_prefix = *ip6_prefix; + d->ip6_prefix_len = ip6_prefix_len; + d->ip4_src = *ip4_src; + d->mtu = mtu; + + if (ip4_prefix_len < 32) + d->shift = 64 - ip6_prefix_len + (32 - ip4_prefix_len); + + /* Create IPv6 route/adjacency */ + fib_prefix_t pfx6 = { + .fp_proto = FIB_PROTOCOL_IP6, + .fp_len = d->ip6_prefix_len, + .fp_addr = { + .ip6 = d->ip6_prefix, + }, + }; + sixrd_dpo_create(DPO_PROTO_IP6, + *sixrd_domain_index, + &dpo_v6); + fib_table_entry_special_dpo_add(0, &pfx6, + FIB_SOURCE_SIXRD, + FIB_ENTRY_FLAG_EXCLUSIVE, + &dpo_v6); + dpo_reset (&dpo_v6); + + /* + * Multiple SIXRD domains may share same source IPv4 TEP + * In this case the route will exist and be SixRD sourced. + * Find the adj (if any) already contributed and modify it + */ + fib_prefix_t pfx4 = { + .fp_proto = FIB_PROTOCOL_IP4, + .fp_len = 32, + .fp_addr = { + .ip4 = d->ip4_src, + }, + }; + fei = fib_table_lookup_exact_match(0, &pfx4); + + if (FIB_NODE_INDEX_INVALID != fei) + { + dpo_id_t dpo = DPO_INVALID; + + if (fib_entry_get_dpo_for_source (fei, FIB_SOURCE_SIXRD, &dpo)) + { + /* + * modify the existing adj to indicate it's shared + * skip to route add. + * It is locked to pair with the unlock below. + */ + const dpo_id_t *sd_dpo; + sixrd_dpo_t *sd; + + ASSERT(DPO_LOAD_BALANCE == dpo.dpoi_type); + + sd_dpo = load_balance_get_bucket(dpo.dpoi_index, 0); + sd = sixrd_dpo_get (sd_dpo->dpoi_index); + + sd->sd_domain = ~0; + dpo_copy (&dpo_v4, sd_dpo); + dpo_reset (&dpo); + + goto route_add; + } + } + /* first time addition of the route */ + sixrd_dpo_create(DPO_PROTO_IP4, + *sixrd_domain_index, + &dpo_v4); + +route_add: + /* + * Create ip4 route. This is a reference counted add. If the prefix + * already exists and is SixRD sourced, it is now SixRD source n+1 times + * and will need to be removed n+1 times. + */ + fib_table_entry_special_dpo_add(0, &pfx4, + FIB_SOURCE_SIXRD, + FIB_ENTRY_FLAG_EXCLUSIVE, + &dpo_v4); + dpo_reset (&dpo_v4); + + return 0; +} + +/* + * sixrd_delete_domain + */ +int +sixrd_delete_domain (u32 sixrd_domain_index) +{ + sixrd_main_t *mm = &sixrd_main; + sixrd_domain_t *d; + + if (pool_is_free_index(mm->domains, sixrd_domain_index)) { + clib_warning("SIXRD domain delete: domain does not exist: %d", + sixrd_domain_index); + return -1; + } + + d = pool_elt_at_index(mm->domains, sixrd_domain_index); + + fib_prefix_t pfx = { + .fp_proto = FIB_PROTOCOL_IP4, + .fp_len = 32, + .fp_addr = { + .ip4 = d->ip4_src, + }, + }; + fib_table_entry_special_remove(0, &pfx, FIB_SOURCE_SIXRD); + + fib_prefix_t pfx6 = { + .fp_proto = FIB_PROTOCOL_IP6, + .fp_len = d->ip6_prefix_len, + .fp_addr = { + .ip6 = d->ip6_prefix, + }, + }; + fib_table_entry_special_remove(0, &pfx6, FIB_SOURCE_SIXRD); + + pool_put(mm->domains, d); + + return 0; +} + +static clib_error_t * +sixrd_add_domain_command_fn (vlib_main_t *vm, + unformat_input_t *input, + vlib_cli_command_t *cmd) +{ + unformat_input_t _line_input, *line_input = &_line_input; + ip4_address_t ip4_prefix; + ip6_address_t ip6_prefix; + ip4_address_t ip4_src; + u32 ip6_prefix_len=0, ip4_prefix_len=0, sixrd_domain_index; + u32 num_m_args = 0; + /* Optional arguments */ + u32 mtu = 0; + + /* Get a line of input. */ + if (!unformat_user(input, unformat_line_input, line_input)) + return 0; + while (unformat_check_input (line_input) != UNFORMAT_END_OF_INPUT) { + if (unformat(line_input, "ip6-pfx %U/%d", unformat_ip6_address, &ip6_prefix, &ip6_prefix_len)) + num_m_args++; + else if (unformat(line_input, "ip4-pfx %U/%d", unformat_ip4_address, &ip4_prefix, &ip4_prefix_len)) + num_m_args++; + else if (unformat(line_input, "ip4-src %U", unformat_ip4_address, &ip4_src)) + num_m_args++; + else if (unformat(line_input, "mtu %d", &mtu)) + num_m_args++; + else + return clib_error_return(0, "unknown input `%U'", + format_unformat_error, input); + } + unformat_free(line_input); + + if (num_m_args < 3) + return clib_error_return(0, "mandatory argument(s) missing"); + + sixrd_create_domain(&ip6_prefix, ip6_prefix_len, &ip4_prefix, ip4_prefix_len, + &ip4_src, &sixrd_domain_index, mtu); + + return 0; +} + +static clib_error_t * +sixrd_del_domain_command_fn (vlib_main_t *vm, + unformat_input_t *input, + vlib_cli_command_t *cmd) +{ + unformat_input_t _line_input, *line_input = &_line_input; + u32 num_m_args = 0; + u32 sixrd_domain_index; + + /* Get a line of input. */ + if (! unformat_user(input, unformat_line_input, line_input)) + return 0; + + while (unformat_check_input(line_input) != UNFORMAT_END_OF_INPUT) { + if (unformat(line_input, "index %d", &sixrd_domain_index)) + num_m_args++; + else + return clib_error_return(0, "unknown input `%U'", + format_unformat_error, input); + } + unformat_free(line_input); + + if (num_m_args != 1) + return clib_error_return(0, "mandatory argument(s) missing"); + + sixrd_delete_domain(sixrd_domain_index); + + return 0; +} + +static u8 * +format_sixrd_domain (u8 *s, va_list *args) +{ + sixrd_domain_t *d = va_arg(*args, sixrd_domain_t *); + sixrd_main_t *mm = &sixrd_main; + + s = format(s, + "[%d] ip6-pfx %U/%d ip4-pfx %U/%d ip4-src %U mtu %d", + d - mm->domains, + format_ip6_address, &d->ip6_prefix, d->ip6_prefix_len, + format_ip4_address, &d->ip4_prefix, d->ip4_prefix_len, + format_ip4_address, &d->ip4_src, d->mtu); + + return s; +} + +static clib_error_t * +show_sixrd_domain_command_fn (vlib_main_t *vm, unformat_input_t *input, vlib_cli_command_t *cmd) +{ + sixrd_main_t *mm = &sixrd_main; + sixrd_domain_t *d; + + if (pool_elts(mm->domains) == 0) + vlib_cli_output(vm, "No SIXRD domains are configured..."); + + pool_foreach(d, mm->domains, ({vlib_cli_output(vm, "%U", format_sixrd_domain, d);})); + + return 0; + +} + +static clib_error_t * +show_sixrd_stats_command_fn (vlib_main_t *vm, unformat_input_t *input, vlib_cli_command_t *cmd) +{ + sixrd_main_t *mm = &sixrd_main; + sixrd_domain_t *d; + int domains = 0, domaincount = 0; + if (pool_elts (mm->domains) == 0) + vlib_cli_output (vm, "No SIXRD domains are configured..."); + + pool_foreach(d, mm->domains, ({ + domains += sizeof(*d); + domaincount++; + })); + + vlib_cli_output(vm, "SIXRD domains structure: %d\n", sizeof (sixrd_domain_t)); + vlib_cli_output(vm, "SIXRD domains: %d (%d bytes)\n", domaincount, domains); + + return 0; +} + +/* + * packet trace format function + */ +u8 * +format_sixrd_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 *); + sixrd_trace_t *t = va_arg (*args, sixrd_trace_t *); + u32 sixrd_domain_index = t->sixrd_domain_index; + + s = format(s, "SIXRD domain index: %d", sixrd_domain_index); + + return s; +} + +VLIB_CLI_COMMAND(sixrd_add_domain_command, static) = { + .path = "sixrd add domain", + .short_help = + "sixrd add domain ip6-pfx <ip6-pfx> ip4-pfx <ip4-pfx> ip4-src <ip4-addr>", + .function = sixrd_add_domain_command_fn, +}; + +VLIB_CLI_COMMAND(sixrd_del_command, static) = { + .path = "sixrd del domain", + .short_help = + "sixrd del domain index <domain>", + .function = sixrd_del_domain_command_fn, +}; + +VLIB_CLI_COMMAND(show_sixrd_domain_command, static) = { + .path = "show sixrd domain", + .function = show_sixrd_domain_command_fn, +}; + +VLIB_CLI_COMMAND(show_sixrd_stats_command, static) = { + .path = "show sixrd stats", + .function = show_sixrd_stats_command_fn, +}; + +/* + * This routine exists to convince the vlib plugin framework that + * we haven't accidentally copied a random .dll into the plugin directory. + * + * Also collects global variable pointers passed from the vpp engine + */ +clib_error_t * +vlib_plugin_register (vlib_main_t * vm, vnet_plugin_handoff_t * h, + int from_early_init) +{ + clib_error_t * error = 0; + sixrd_main_t *mm = &sixrd_main; + + mm->vnet_main = vnet_get_main(); + mm->vlib_main = vm; + + return error; +} + +static clib_error_t * sixrd_init (vlib_main_t * vm) +{ + sixrd_dpo_module_init (); + + return (NULL); +} + +VLIB_INIT_FUNCTION (sixrd_init); diff --git a/src/plugins/sixrd/sixrd.h b/src/plugins/sixrd/sixrd.h new file mode 100644 index 00000000000..56714c9e3bd --- /dev/null +++ b/src/plugins/sixrd/sixrd.h @@ -0,0 +1,141 @@ +/*--------------------------------------------------------------------------- + * Copyright (c) 2009-2014 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 <stdbool.h> +#include <vppinfra/error.h> +#include <vnet/vnet.h> +#include <vnet/ip/ip.h> +#include <vnet/fib/ip6_fib.h> + +#include "sixrd_dpo.h" + +int sixrd_create_domain(ip6_address_t *ip6_prefix, u8 ip6_prefix_len, + ip4_address_t *ip4_prefix, u8 ip4_prefix_len, + ip4_address_t *ip4_src, u32 *sixrd_domain_index, u16 mtu); +int sixrd_delete_domain(u32 sixrd_domain_index); +u8 *format_sixrd_trace(u8 *s, va_list *args); + +typedef struct { + ip6_address_t ip6_prefix; + ip4_address_t ip4_prefix; + ip4_address_t ip4_src; + u8 ip6_prefix_len; + u8 ip4_prefix_len; + + /* helpers */ + u8 shift; + + u16 mtu; +} sixrd_domain_t; + +typedef struct { + /* pool of SIXRD domains */ + sixrd_domain_t *domains; + + /* convenience */ + vlib_main_t *vlib_main; + vnet_main_t *vnet_main; +} sixrd_main_t; + +#define foreach_sixrd_error \ + /* Must be first. */ \ + _(NONE, "valid SIXRD packets") \ + _(BAD_PROTOCOL, "bad protocol") \ + _(WRONG_ICMP_TYPE, "wrong icmp type") \ + _(SEC_CHECK, "security check failed") \ + _(ICMP, "unable to translate ICMP") \ + _(UNKNOWN, "unknown") \ + _(NO_DOMAIN, "no domain") \ + _(ENCAPSULATED, "encapsulated") \ + _(DECAPSULATED, "decapsulated") \ + _(TRANSLATED_4TO6, "translated 4 to 6") \ + _(TRANSLATED_6TO4, "translated 6 to 4") \ + _(FRAGMENT, "fragment handling error") \ + _(FRAGMENT_QUEUED, "dropped, missing first fragment") \ + _(FRAGMENTED, "packets requiring fragmentation") \ + _(FRAGMENT_PARTS, "fragment parts") \ + _(MALFORMED, "malformed packet") + +typedef enum { +#define _(sym,str) SIXRD_ERROR_##sym, + foreach_sixrd_error +#undef _ + SIXRD_N_ERROR, + } sixrd_error_t; + +typedef struct { + u32 sixrd_domain_index; +} sixrd_trace_t; + +sixrd_main_t sixrd_main; + +/* + * sixrd_get_addr + */ +static_always_inline u32 +sixrd_get_addr (sixrd_domain_t *d, u64 dal) +{ + + /* 1:1 mode */ + if (d->ip4_prefix_len == 32) return (d->ip4_prefix.as_u32); + + /* Grab 32 - ip4_prefix_len bits out of IPv6 address from offset ip6_prefix_len */ + return (d->ip4_prefix.as_u32 | (u32)(dal >> d->shift)); +} + +/* + * Get the SIXRD domain from an IPv6 lookup adjacency. + */ +static_always_inline sixrd_domain_t * +ip6_sixrd_get_domain (u32 sdi, u32 *sixrd_domain_index) +{ + sixrd_main_t *mm = &sixrd_main; + sixrd_dpo_t *sd; + + sd = sixrd_dpo_get(sdi); + + ASSERT(sd); + *sixrd_domain_index = sd->sd_domain; + return pool_elt_at_index(mm->domains, *sixrd_domain_index); +} + +/* + * Get the SIXRD domain from an IPv4 lookup adjacency. + * If the IPv4 address is not shared, no lookup is required. + * The IPv6 address is used otherwise. + */ +static_always_inline sixrd_domain_t * +ip4_sixrd_get_domain (u32 sdi, ip6_address_t *addr, + u32 *sixrd_domain_index, u8 *error) +{ + sixrd_main_t *mm = &sixrd_main; + sixrd_dpo_t *sd; + + sd = sixrd_dpo_get(sdi); + *sixrd_domain_index = sd->sd_domain; + if (*sixrd_domain_index != ~0) + return pool_elt_at_index(mm->domains, *sixrd_domain_index); + + u32 lbi = ip6_fib_table_fwding_lookup(&ip6_main, 0, addr); + const dpo_id_t *dpo = load_balance_get_bucket(lbi, 0); + if (PREDICT_TRUE(dpo->dpoi_type == sixrd_dpo_type)) + { + sd = sixrd_dpo_get(dpo->dpoi_index); + *sixrd_domain_index = sd->sd_domain; + return pool_elt_at_index(mm->domains, *sixrd_domain_index); + } + *error = SIXRD_ERROR_NO_DOMAIN; + return NULL; +} diff --git a/src/plugins/sixrd/sixrd_dpo.c b/src/plugins/sixrd/sixrd_dpo.c new file mode 100644 index 00000000000..88a079350a3 --- /dev/null +++ b/src/plugins/sixrd/sixrd_dpo.c @@ -0,0 +1,132 @@ +/* + * 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: + * + * 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 "sixrd_dpo.h" +#include <vnet/ip/ip.h> + +/** + * pool of all MPLS Label DPOs + */ +sixrd_dpo_t *sixrd_dpo_pool; + +/** + * The register SIXRD DPO type + */ +dpo_type_t sixrd_dpo_type; + +static sixrd_dpo_t * +sixrd_dpo_alloc (void) +{ + sixrd_dpo_t *sd; + + pool_get_aligned(sixrd_dpo_pool, sd, CLIB_CACHE_LINE_BYTES); + memset(sd, 0, sizeof(*sd)); + + return (sd); +} + +static index_t +sixrd_dpo_get_index (sixrd_dpo_t *sd) +{ + return (sd - sixrd_dpo_pool); +} + +void +sixrd_dpo_create (dpo_proto_t dproto, + u32 domain_index, + dpo_id_t *dpo) +{ + sixrd_dpo_t *sd; + + sd = sixrd_dpo_alloc(); + sd->sd_domain = domain_index; + sd->sd_proto = dproto; + + dpo_set(dpo, + sixrd_dpo_type, + dproto, + sixrd_dpo_get_index(sd)); +} + +u8* +format_sixrd_dpo (u8 *s, va_list *args) +{ + index_t index = va_arg (*args, index_t); + CLIB_UNUSED(u32 indent) = va_arg (*args, u32); + sixrd_dpo_t *sd; + + sd = sixrd_dpo_get(index); + + return (format(s, "sixrd:[%d]:%U domain:%d", + index, + format_dpo_proto, sd->sd_proto, + sd->sd_domain)); +} + + +static void +sixrd_dpo_lock (dpo_id_t *dpo) +{ + sixrd_dpo_t *sd; + + sd = sixrd_dpo_get(dpo->dpoi_index); + + sd->sd_locks++; +} + +static void +sixrd_dpo_unlock (dpo_id_t *dpo) +{ + sixrd_dpo_t *sd; + + sd = sixrd_dpo_get(dpo->dpoi_index); + + sd->sd_locks--; + + if (0 == sd->sd_locks) + { + pool_put(sixrd_dpo_pool, sd); + } +} + +const static dpo_vft_t sd_vft = { + .dv_lock = sixrd_dpo_lock, + .dv_unlock = sixrd_dpo_unlock, + .dv_format = format_sixrd_dpo, +}; + +const static char* const sixrd_ip4_nodes[] = +{ + "ip4-sixrd", + NULL, +}; +const static char* const sixrd_ip6_nodes[] = +{ + "ip6-sixrd", + NULL, +}; + +const static char* const * const sixrd_nodes[DPO_PROTO_NUM] = +{ + [DPO_PROTO_IP4] = sixrd_ip4_nodes, + [DPO_PROTO_IP6] = sixrd_ip6_nodes, + [DPO_PROTO_MPLS] = NULL, +}; + +void +sixrd_dpo_module_init (void) +{ + sixrd_dpo_type = dpo_register_new_type(&sd_vft, sixrd_nodes); +} diff --git a/src/plugins/sixrd/sixrd_dpo.h b/src/plugins/sixrd/sixrd_dpo.h new file mode 100644 index 00000000000..17142288451 --- /dev/null +++ b/src/plugins/sixrd/sixrd_dpo.h @@ -0,0 +1,61 @@ +/* + * 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: + * + * 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. + */ + +#ifndef __SIXRD_DPO_H__ +#define __SIXRD_DPO_H__ + +#include <vnet/vnet.h> +#include <vnet/dpo/dpo.h> + +/** + * A representation of a 6RD DPO + */ +typedef struct sixrd_dpo_t +{ + /** + * The dat-plane protocol + */ + dpo_proto_t sd_proto; + + /** + * the SIXRD domain index + */ + u32 sd_domain; + + /** + * Number of locks/users of the label + */ + u16 sd_locks; +} sixrd_dpo_t; + +extern void sixrd_dpo_create (dpo_proto_t dproto, + u32 domain_index, + dpo_id_t *dpo); + +/* + * Encapsulation violation for fast data-path access + */ +extern sixrd_dpo_t *sixrd_dpo_pool; +extern dpo_type_t sixrd_dpo_type; + +static inline sixrd_dpo_t * +sixrd_dpo_get (index_t index) +{ + return (pool_elt_at_index(sixrd_dpo_pool, index)); +} + +extern void sixrd_dpo_module_init(void); + +#endif |