1.. BSD LICENSE 2 Copyright(c) 2010-2014 Intel Corporation. All rights reserved. 3 All rights reserved. 4 5 Redistribution and use in source and binary forms, with or without 6 modification, are permitted provided that the following conditions 7 are met: 8 9 * Redistributions of source code must retain the above copyright 10 notice, this list of conditions and the following disclaimer. 11 * Redistributions in binary form must reproduce the above copyright 12 notice, this list of conditions and the following disclaimer in 13 the documentation and/or other materials provided with the 14 distribution. 15 * Neither the name of Intel Corporation nor the names of its 16 contributors may be used to endorse or promote products derived 17 from this software without specific prior written permission. 18 19 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 20 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT 21 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR 22 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT 23 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, 24 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT 25 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 26 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 27 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 28 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE 29 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 30 31Distributor Sample Application 32============================== 33 34The distributor sample application is a simple example of packet distribution 35to cores using the Data Plane Development Kit (DPDK). 36 37Overview 38-------- 39 40The distributor application performs the distribution of packets that are received 41on an RX_PORT to different cores. When processed by the cores, the destination 42port of a packet is the port from the enabled port mask adjacent to the one on 43which the packet was received, that is, if the first four ports are enabled 44(port mask 0xf), ports 0 and 1 RX/TX into each other, and ports 2 and 3 RX/TX 45into each other. 46 47This application can be used to benchmark performance using the traffic 48generator as shown in the figure below. 49 50.. _figure_dist_perf: 51 52.. figure:: img/dist_perf.* 53 54 Performance Benchmarking Setup (Basic Environment) 55 56Compiling the Application 57------------------------- 58 59To compile the sample application see :doc:`compiling`. 60 61The application is located in the ``distributor`` sub-directory. 62 63Running the Application 64----------------------- 65 66#. The application has a number of command line options: 67 68 .. code-block:: console 69 70 ./build/distributor_app [EAL options] -- -p PORTMASK 71 72 where, 73 74 * -p PORTMASK: Hexadecimal bitmask of ports to configure 75 76#. To run the application in linuxapp environment with 10 lcores, 4 ports, 77 issue the command: 78 79 .. code-block:: console 80 81 $ ./build/distributor_app -l 1-9,22 -n 4 -- -p f 82 83#. Refer to the DPDK Getting Started Guide for general information on running 84 applications and the Environment Abstraction Layer (EAL) options. 85 86Explanation 87----------- 88 89The distributor application consists of four types of threads: a receive 90thread (``lcore_rx()``), a distributor thread (``lcore_dist()``), a set of 91worker threads (``lcore_worker()``), and a transmit thread(``lcore_tx()``). 92How these threads work together is shown in :numref:`figure_dist_app` below. 93The ``main()`` function launches threads of these four types. Each thread 94has a while loop which will be doing processing and which is terminated 95only upon SIGINT or ctrl+C. 96 97The receive thread receives the packets using ``rte_eth_rx_burst()`` and will 98enqueue them to an rte_ring. The distributor thread will dequeue the packets 99from the ring and assign them to workers (using ``rte_distributor_process()`` API). 100This assignment is based on the tag (or flow ID) of the packet - indicated by 101the hash field in the mbuf. For IP traffic, this field is automatically filled 102by the NIC with the "usr" hash value for the packet, which works as a per-flow 103tag. The distributor thread communicates with the worker threads using a 104cache-line swapping mechanism, passing up to 8 mbuf pointers at a time 105(one cache line) to each worker. 106 107More than one worker thread can exist as part of the application, and these 108worker threads do simple packet processing by requesting packets from 109the distributor, doing a simple XOR operation on the input port mbuf field 110(to indicate the output port which will be used later for packet transmission) 111and then finally returning the packets back to the distributor thread. 112 113The distributor thread will then call the distributor api 114``rte_distributor_returned_pkts()`` to get the processed packets, and will enqueue 115them to another rte_ring for transfer to the TX thread for transmission on the 116output port. The transmit thread will dequeue the packets from the ring and 117transmit them on the output port specified in packet mbuf. 118 119Users who wish to terminate the running of the application have to press ctrl+C 120(or send SIGINT to the app). Upon this signal, a signal handler provided 121in the application will terminate all running threads gracefully and print 122final statistics to the user. 123 124.. _figure_dist_app: 125 126.. figure:: img/dist_app.* 127 128 Distributor Sample Application Layout 129 130 131Debug Logging Support 132--------------------- 133 134Debug logging is provided as part of the application; the user needs to uncomment 135the line "#define DEBUG" defined in start of the application in main.c to enable debug logs. 136 137Statistics 138---------- 139 140The main function will print statistics on the console every second. These 141statistics include the number of packets enqueued and dequeued at each stage 142in the application, and also key statistics per worker, including how many 143packets of each burst size (1-8) were sent to each worker thread. 144 145Application Initialization 146-------------------------- 147 148Command line parsing is done in the same way as it is done in the L2 Forwarding Sample 149Application. See :ref:`l2_fwd_app_cmd_arguments`. 150 151Mbuf pool initialization is done in the same way as it is done in the L2 Forwarding 152Sample Application. See :ref:`l2_fwd_app_mbuf_init`. 153 154Driver Initialization is done in same way as it is done in the L2 Forwarding Sample 155Application. See :ref:`l2_fwd_app_dvr_init`. 156 157RX queue initialization is done in the same way as it is done in the L2 Forwarding 158Sample Application. See :ref:`l2_fwd_app_rx_init`. 159 160TX queue initialization is done in the same way as it is done in the L2 Forwarding 161Sample Application. See :ref:`l2_fwd_app_tx_init`. 162