From patchwork Thu Oct 26 09:09:17 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Iremonger, Bernard" X-Patchwork-Id: 30938 Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id F1E5F1BA88; Thu, 26 Oct 2017 11:09:28 +0200 (CEST) Received: from mga03.intel.com (mga03.intel.com [134.134.136.65]) by dpdk.org (Postfix) with ESMTP id A15C21BA78 for ; Thu, 26 Oct 2017 11:09:26 +0200 (CEST) Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by orsmga103.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 26 Oct 2017 02:09:26 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.43,434,1503385200"; d="scan'208";a="167840368" Received: from sivswdev01.ir.intel.com (HELO localhost.localdomain) ([10.237.217.45]) by fmsmga006.fm.intel.com with ESMTP; 26 Oct 2017 02:09:25 -0700 From: Bernard Iremonger To: dev@dpdk.org, john.mcnamara@intel.com Cc: Bernard Iremonger Date: Thu, 26 Oct 2017 10:09:17 +0100 Message-Id: <1509008957-11812-3-git-send-email-bernard.iremonger@intel.com> X-Mailer: git-send-email 1.7.0.7 In-Reply-To: <1509008957-11812-1-git-send-email-bernard.iremonger@intel.com> References: <1509008957-11812-1-git-send-email-bernard.iremonger@intel.com> Subject: [dpdk-dev] [PATCH 2/2] doc: flow classify sample app guide X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" This file documents the flow_classify sample application which is used to demonstate use of the Flow Classify library, librte_flow_classify. Updated MAINTAINERS file Signed-off-by: Bernard Iremonger --- MAINTAINERS | 1 + doc/guides/sample_app_ug/flow_classify.rst | 592 +++++++++++++++++++++++++++++ doc/guides/sample_app_ug/index.rst | 1 + 3 files changed, 594 insertions(+) create mode 100644 doc/guides/sample_app_ug/flow_classify.rst diff --git a/MAINTAINERS b/MAINTAINERS index 117d104..63ba5ab 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -719,6 +719,7 @@ F: lib/librte_flow_classify/ F: test/test/test_flow_classify* F: examples/flow_classify/ F: doc/guides/prog_guide/flow_classify_lib.rst +F: doc/guides/sample_app_ug/flow_classify.rst Distributor M: Bruce Richardson diff --git a/doc/guides/sample_app_ug/flow_classify.rst b/doc/guides/sample_app_ug/flow_classify.rst new file mode 100644 index 0000000..b2f5e4f --- /dev/null +++ b/doc/guides/sample_app_ug/flow_classify.rst @@ -0,0 +1,592 @@ +.. BSD LICENSE + Copyright(c) 2017 Intel Corporation. All rights reserved. + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in + the documentation and/or other materials provided with the + distribution. + * Neither the name of Intel Corporation nor the names of its + contributors may be used to endorse or promote products derived + from this software without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +Flow Classify Sample Application +================================ + +The Flow Classify sample application is based on the simple *skeleton* example +of a forwarding application. + +It is intended as a demonstration of the basic components of a DPDK forwarding +application which uses the Flow Classify library API's. + +Please refer to the +:doc:`../prog_guide/flow_classify_lib` +for more information. + +Compiling the Application +------------------------- + +To compile the application export the path to the DPDK source tree and go to +the example directory: + +.. code-block:: console + + export RTE_SDK=/path/to/rte_sdk + + cd ${RTE_SDK}/examples/flow_classify + +Set the target, for example: + +.. code-block:: console + + export RTE_TARGET=x86_64-native-linuxapp-gcc + +See the *DPDK Getting Started* Guide for possible ``RTE_TARGET`` values. + +Build the application as follows: + +.. code-block:: console + + make clean + + make + +Running the Application +----------------------- + +To run the example in a ``linuxapp`` environment: + +.. code-block:: console + + ./build/flow_classify -c 4 -n 4 -- --rule_ipv4="../ipv4_rules_file.txt" + +Refer to *DPDK Getting Started Guide* for general information on running +applications and the Environment Abstraction Layer (EAL) options. + +Sample ipv4_rules_file.txt +-------------------------- + +.. code-block:: console + + #file format: + #src_ip/masklen dst_ip/masklen src_port : mask dst_port : mask proto/mask priority + # + 2.2.2.3/24 2.2.2.7/24 32 : 0xffff 33 : 0xffff 17/0xff 0 + 9.9.9.3/24 9.9.9.7/24 32 : 0xffff 33 : 0xffff 17/0xff 1 + 9.9.9.3/24 9.9.9.7/24 32 : 0xffff 33 : 0xffff 6/0xff 2 + 9.9.8.3/24 9.9.8.7/24 32 : 0xffff 33 : 0xffff 6/0xff 3 + 6.7.8.9/24 2.3.4.5/24 32 : 0x0000 33 : 0x0000 132/0xff 4 + +Explanation +----------- + +The following sections provide an explanation of the main components of the +code. + +All DPDK library functions used in the sample code are prefixed with ``rte_`` +and are explained in detail in the *DPDK API Documentation*. + +ACL field definitions for the IPv4 5 tuple rule +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +The following field definitions are used when creating the ACL table during +initialisation of the ``Flow Classify`` application.. + +.. code-block:: c + + enum { + PROTO_FIELD_IPV4, + SRC_FIELD_IPV4, + DST_FIELD_IPV4, + SRCP_FIELD_IPV4, + DSTP_FIELD_IPV4, + NUM_FIELDS_IPV4 + }; + + enum { + PROTO_INPUT_IPV4, + SRC_INPUT_IPV4, + DST_INPUT_IPV4, + SRCP_DESTP_INPUT_IPV4 + }; + + static struct rte_acl_field_def ipv4_defs[NUM_FIELDS_IPV4] = { + /* first input field - always one byte long. */ + { + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint8_t), + .field_index = PROTO_FIELD_IPV4, + .input_index = PROTO_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + offsetof(struct ipv4_hdr, next_proto_id), + }, + /* next input field (IPv4 source address) - 4 consecutive bytes. */ + { + /* rte_flow uses a bit mask for IPv4 addresses */ + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint32_t), + .field_index = SRC_FIELD_IPV4, + .input_index = SRC_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + offsetof(struct ipv4_hdr, src_addr), + }, + /* next input field (IPv4 destination address) - 4 consecutive bytes. */ + { + /* rte_flow uses a bit mask for IPv4 addresses */ + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint32_t), + .field_index = DST_FIELD_IPV4, + .input_index = DST_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + offsetof(struct ipv4_hdr, dst_addr), + }, + /* + * Next 2 fields (src & dst ports) form 4 consecutive bytes. + * They share the same input index. + */ + { + /* rte_flow uses a bit mask for protocol ports */ + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint16_t), + .field_index = SRCP_FIELD_IPV4, + .input_index = SRCP_DESTP_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + sizeof(struct ipv4_hdr) + + offsetof(struct tcp_hdr, src_port), + }, + { + /* rte_flow uses a bit mask for protocol ports */ + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint16_t), + .field_index = DSTP_FIELD_IPV4, + .input_index = SRCP_DESTP_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + sizeof(struct ipv4_hdr) + + offsetof(struct tcp_hdr, dst_port), + }, + }; + +The Main Function +~~~~~~~~~~~~~~~~~ + +The ``main()`` function performs the initialization and calls the execution +threads for each lcore. + +The first task is to initialize the Environment Abstraction Layer (EAL). +The ``argc`` and ``argv`` arguments are provided to the ``rte_eal_init()`` +function. The value returned is the number of parsed arguments: + +.. code-block:: c + + int ret = rte_eal_init(argc, argv); + if (ret < 0) + rte_exit(EXIT_FAILURE, "Error with EAL initialization\n"); + +It then parses the flow_classify application arguments + +.. code-block:: c + + ret = parse_args(argc, argv); + if (ret < 0) + rte_exit(EXIT_FAILURE, "Invalid flow_classify parameters\n"); + +The ``main()`` function also allocates a mempool to hold the mbufs +(Message Buffers) used by the application: + +.. code-block:: c + + mbuf_pool = rte_mempool_create("MBUF_POOL", + NUM_MBUFS * nb_ports, + MBUF_SIZE, + MBUF_CACHE_SIZE, + sizeof(struct rte_pktmbuf_pool_private), + rte_pktmbuf_pool_init, NULL, + rte_pktmbuf_init, NULL, + rte_socket_id(), + 0); + +mbufs are the packet buffer structure used by DPDK. They are explained in +detail in the "Mbuf Library" section of the *DPDK Programmer's Guide*. + +The ``main()`` function also initializes all the ports using the user defined +``port_init()`` function which is explained in the next section: + +.. code-block:: c + + for (portid = 0; portid < nb_ports; portid++) { + if (port_init(portid, mbuf_pool) != 0) { + rte_exit(EXIT_FAILURE, + "Cannot init port %" PRIu8 "\n", portid); + } + } + +The ``main()`` function creates the ``flow classifier object`` and adds an ``ACL +table`` to the flow classifer. + +.. code-block:: c + + struct flow_classifier { + struct rte_flow_classifier *cls; + uint32_t table_id[RTE_FLOW_CLASSIFY_TABLE_MAX]; + }; + + struct flow_classifier_acl { + struct flow_classifier cls; + } __rte_cache_aligned; + + /* Memory allocation */ + size = RTE_CACHE_LINE_ROUNDUP(sizeof(struct flow_classifier_acl)); + cls_app = rte_zmalloc(NULL, size, RTE_CACHE_LINE_SIZE); + if (cls_app == NULL) + rte_exit(EXIT_FAILURE, "Cannot allocate classifier memory\n"); + + cls_params.name = "flow_classifier"; + cls_params.socket_id = socket_id; + cls_params.type = RTE_FLOW_CLASSIFY_TABLE_TYPE_ACL; + + cls_app->cls = rte_flow_classifier_create(&cls_params); + if (cls_app->cls == NULL) { + rte_free(cls_app); + rte_exit(EXIT_FAILURE, "Cannot create classifier\n"); + } + + /* initialise ACL table params */ + table_acl_params.name = "table_acl_ipv4_5tuple"; + table_acl_params.n_rule_fields = RTE_DIM(ipv4_defs); + table_acl_params.n_rules = FLOW_CLASSIFY_MAX_RULE_NUM; + memcpy(table_acl_params.field_format, ipv4_defs, sizeof(ipv4_defs)); + + /* initialise table create params */ + cls_table_params.ops = &rte_table_acl_ops, + cls_table_params.arg_create = &table_acl_params, + cls_table_params.table_metadata_size = 0; + + ret = rte_flow_classify_table_create(cls_app->cls, &cls_table_params, + &cls->table_id[0]); + if (ret) { + rte_flow_classifier_free(cls_app->cls); + rte_free(cls); + rte_exit(EXIT_FAILURE, "Failed to create classifier table\n"); + } + +It then reads the ipv4_rules_file.txt file and initialises the parameters for +the ``rte_flow_classify_table_entry_add`` API. +This API adds a rule to the ACL table. + +.. code-block:: c + + if (add_rules(parm_config.rule_ipv4_name)) { + rte_flow_classifier_free(cls_app->cls); + rte_free(cls_app); + rte_exit(EXIT_FAILURE, "Failed to add rules\n"); + } + +Once the initialization is complete, the application is ready to launch a +function on an lcore. In this example ``lcore_main()`` is called on a single +lcore. + +.. code-block:: c + + lcore_main(cls_app); + +The ``lcore_main()`` function is explained below. + +The Port Initialization Function +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +The main functional part of the port initialization used in the Basic +Forwarding application is shown below: + +.. code-block:: c + + static inline int + port_init(uint8_t port, struct rte_mempool *mbuf_pool) + { + struct rte_eth_conf port_conf = port_conf_default; + const uint16_t rx_rings = 1, tx_rings = 1; + struct ether_addr addr; + int retval; + uint16_t q; + + if (port >= rte_eth_dev_count()) + return -1; + + /* Configure the Ethernet device. */ + retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf); + if (retval != 0) + return retval; + + /* Allocate and set up 1 RX queue per Ethernet port. */ + for (q = 0; q < rx_rings; q++) { + retval = rte_eth_rx_queue_setup(port, q, RX_RING_SIZE, + rte_eth_dev_socket_id(port), NULL, mbuf_pool); + if (retval < 0) + return retval; + } + + /* Allocate and set up 1 TX queue per Ethernet port. */ + for (q = 0; q < tx_rings; q++) { + retval = rte_eth_tx_queue_setup(port, q, TX_RING_SIZE, + rte_eth_dev_socket_id(port), NULL); + if (retval < 0) + return retval; + } + + /* Start the Ethernet port. */ + retval = rte_eth_dev_start(port); + if (retval < 0) + return retval; + + /* Display the port MAC address. */ + rte_eth_macaddr_get(port, &addr); + printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8 + " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n", + port, + addr.addr_bytes[0], addr.addr_bytes[1], + addr.addr_bytes[2], addr.addr_bytes[3], + addr.addr_bytes[4], addr.addr_bytes[5]); + + /* Enable RX in promiscuous mode for the Ethernet device. */ + rte_eth_promiscuous_enable(port); + + return 0; + } + +The Ethernet ports are configured with default settings using the +``rte_eth_dev_configure()`` function and the ``port_conf_default`` struct. + +.. code-block:: c + + static const struct rte_eth_conf port_conf_default = { + .rxmode = { .max_rx_pkt_len = ETHER_MAX_LEN } + }; + +For this example the ports are set up with 1 RX and 1 TX queue using the +``rte_eth_rx_queue_setup()`` and ``rte_eth_tx_queue_setup()`` functions. + +The Ethernet port is then started: + +.. code-block:: c + + retval = rte_eth_dev_start(port); + + +Finally the RX port is set in promiscuous mode: + +.. code-block:: c + + rte_eth_promiscuous_enable(port); + +The Add Rules function +~~~~~~~~~~~~~~~~~~~~~~ + +The ``add_rules`` function reads the ``ipv4_rules_file.txt`` file and calls the +``add_classify_rule`` function which calls the +``rte_flow_classify_table_entry_add`` API. + +.. code-block:: c + + static int + add_rules(const char *rule_path) + { + FILE *fh; + char buff[LINE_MAX]; + unsigned int i = 0; + unsigned int total_num = 0; + struct rte_eth_ntuple_filter ntuple_filter; + + fh = fopen(rule_path, "rb"); + if (fh == NULL) + rte_exit(EXIT_FAILURE, "%s: Open %s failed\n", __func__, + rule_path); + + fseek(fh, 0, SEEK_SET); + + i = 0; + while (fgets(buff, LINE_MAX, fh) != NULL) { + i++; + + if (is_bypass_line(buff)) + continue; + + if (total_num >= FLOW_CLASSIFY_MAX_RULE_NUM - 1) { + printf("\nINFO: classify rule capacity %d reached\n", + total_num); + break; + } + + if (parse_ipv4_5tuple_rule(buff, &ntuple_filter) != 0) + rte_exit(EXIT_FAILURE, + "%s Line %u: parse rules error\n", + rule_path, i); + + if (add_classify_rule(&ntuple_filter) != 0) + rte_exit(EXIT_FAILURE, "add rule error\n"); + + total_num++; + } + + fclose(fh); + return 0; + } + + +The Lcore Main function +~~~~~~~~~~~~~~~~~~~~~~~ + +As we saw above the ``main()`` function calls an application function on the +available lcores. +The ``lcore_main`` function calls the ``rte_flow_classifier_query`` API. +For the Basic Forwarding application the ``lcore_main`` function looks like the +following: + +.. code-block:: c + + /* flow classify data */ + static int num_classify_rules; + static struct rte_flow_classify_rule *rules[MAX_NUM_CLASSIFY]; + static struct rte_flow_classify_ipv4_5tuple_stats ntuple_stats; + static struct rte_flow_classify_stats classify_stats = { + .stats = (void *)&ntuple_stats + }; + + static __attribute__((noreturn)) void + lcore_main(cls_app) + { + const uint8_t nb_ports = rte_eth_dev_count(); + uint8_t port; + + /* + * Check that the port is on the same NUMA node as the polling thread + * for best performance. + */ + for (port = 0; port < nb_ports; port++) + if (rte_eth_dev_socket_id(port) > 0 && + rte_eth_dev_socket_id(port) != (int)rte_socket_id()) { + printf("\n\n"); + printf("WARNING: port %u is on remote NUMA node\n", + port); + printf("to polling thread.\n"); + printf("Performance will not be optimal.\n"); + + printf("\nCore %u forwarding packets. \n", + rte_lcore_id()); + printf("[Ctrl+C to quit]\n + } + + /* Run until the application is quit or killed. */ + for (;;) { + /* + * Receive packets on a port and forward them on the paired + * port. The mapping is 0 -> 1, 1 -> 0, 2 -> 3, 3 -> 2, etc. + */ + for (port = 0; port < nb_ports; port++) { + + /* Get burst of RX packets, from first port of pair. */ + struct rte_mbuf *bufs[BURST_SIZE]; + const uint16_t nb_rx = rte_eth_rx_burst(port, 0, + bufs, BURST_SIZE); + + if (unlikely(nb_rx == 0)) + continue; + + for (i = 0; i < MAX_NUM_CLASSIFY; i++) { + if (rules[i]) { + ret = rte_flow_classifier_query( + cls_app->cls, + cls_app->table_id[0], + bufs, nb_rx, rules[i], + &classify_stats); + if (ret) + printf( + "rule [%d] query failed ret [%d]\n\n", + i, ret); + else { + printf( + "rule[%d] count=%"PRIu64"\n", + i, ntuple_stats.counter1); + + printf("proto = %d\n", + ntuple_stats.ipv4_5tuple.proto); + } + } + } + + /* Send burst of TX packets, to second port of pair. */ + const uint16_t nb_tx = rte_eth_tx_burst(port ^ 1, 0, + bufs, nb_rx); + + /* Free any unsent packets. */ + if (unlikely(nb_tx < nb_rx)) { + uint16_t buf; + for (buf = nb_tx; buf < nb_rx; buf++) + rte_pktmbuf_free(bufs[buf]); + } + } + } + } + +The main work of the application is done within the loop: + +.. code-block:: c + + for (;;) { + for (port = 0; port < nb_ports; port++) { + + /* Get burst of RX packets, from first port of pair. */ + struct rte_mbuf *bufs[BURST_SIZE]; + const uint16_t nb_rx = rte_eth_rx_burst(port, 0, + bufs, BURST_SIZE); + + if (unlikely(nb_rx == 0)) + continue; + + /* Send burst of TX packets, to second port of pair. */ + const uint16_t nb_tx = rte_eth_tx_burst(port ^ 1, 0, + bufs, nb_rx); + + /* Free any unsent packets. */ + if (unlikely(nb_tx < nb_rx)) { + uint16_t buf; + for (buf = nb_tx; buf < nb_rx; buf++) + rte_pktmbuf_free(bufs[buf]); + } + } + } + +Packets are received in bursts on the RX ports and transmitted in bursts on +the TX ports. The ports are grouped in pairs with a simple mapping scheme +using the an XOR on the port number:: + + 0 -> 1 + 1 -> 0 + + 2 -> 3 + 3 -> 2 + + etc. + +The ``rte_eth_tx_burst()`` function frees the memory buffers of packets that +are transmitted. If packets fail to transmit, ``(nb_tx < nb_rx)``, then they +must be freed explicitly using ``rte_pktmbuf_free()``. + +The forwarding loop can be interrupted and the application closed using +``Ctrl-C``. diff --git a/doc/guides/sample_app_ug/index.rst b/doc/guides/sample_app_ug/index.rst index 0c17ebb..6e66046 100644 --- a/doc/guides/sample_app_ug/index.rst +++ b/doc/guides/sample_app_ug/index.rst @@ -78,6 +78,7 @@ Sample Applications User Guides ptpclient performance_thread ipsec_secgw + flow_classify **Figures**