1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2015 Intel Corporation.
4 PTP Client Sample Application
5 =============================
7 The PTP (Precision Time Protocol) client sample application is a simple
8 example of using the DPDK IEEE1588 API to communicate with a PTP master clock
9 to synchronize the time on the NIC and, optionally, on the Linux system.
11 Note, PTP is a time syncing protocol and cannot be used within DPDK as a
12 time-stamping mechanism. See the following for an explanation of the protocol:
13 `Precision Time Protocol
14 <https://en.wikipedia.org/wiki/Precision_Time_Protocol>`_.
20 The PTP sample application is intended as a simple reference implementation of
21 a PTP client using the DPDK IEEE1588 API.
22 In order to keep the application simple the following assumptions are made:
24 * The first discovered master is the master for the session.
25 * Only L2 PTP packets are supported.
26 * Only the PTP v2 protocol is supported.
27 * Only the slave clock is implemented.
30 How the Application Works
31 -------------------------
33 .. _figure_ptpclient_highlevel:
35 .. figure:: img/ptpclient.*
37 PTP Synchronization Protocol
39 The PTP synchronization in the sample application works as follows:
41 * Master sends *Sync* message - the slave saves it as T2.
42 * Master sends *Follow Up* message and sends time of T1.
43 * Slave sends *Delay Request* frame to PTP Master and stores T3.
44 * Master sends *Delay Response* T4 time which is time of received T3.
46 The adjustment for slave can be represented as:
48 adj = -[(T2-T1)-(T4 - T3)]/2
50 If the command line parameter ``-T 1`` is used the application also
51 synchronizes the PTP PHC clock with the Linux kernel clock.
53 Compiling the Application
54 -------------------------
56 To compile the sample application see :doc:`compiling`.
58 The application is located in the ``ptpclient`` sub-directory.
61 To compile the application edit the ``config/common_linux`` configuration file to enable IEEE1588
62 and then recompile DPDK:
64 .. code-block:: console
66 CONFIG_RTE_LIBRTE_IEEE1588=y
68 Running the Application
69 -----------------------
71 To run the example in a ``linux`` environment:
73 .. code-block:: console
75 ./build/ptpclient -l 1 -n 4 -- -p 0x1 -T 0
77 Refer to *DPDK Getting Started Guide* for general information on running
78 applications and the Environment Abstraction Layer (EAL) options.
80 * ``-p portmask``: Hexadecimal portmask.
81 * ``-T 0``: Update only the PTP slave clock.
82 * ``-T 1``: Update the PTP slave clock and synchronize the Linux Kernel to the PTP clock.
88 The following sections provide an explanation of the main components of the
91 All DPDK library functions used in the sample code are prefixed with ``rte_``
92 and are explained in detail in the *DPDK API Documentation*.
98 The ``main()`` function performs the initialization and calls the execution
99 threads for each lcore.
101 The first task is to initialize the Environment Abstraction Layer (EAL). The
102 ``argc`` and ``argv`` arguments are provided to the ``rte_eal_init()``
103 function. The value returned is the number of parsed arguments:
107 int ret = rte_eal_init(argc, argv);
109 rte_exit(EXIT_FAILURE, "Error with EAL initialization\n");
111 And than we parse application specific arguments
118 ret = ptp_parse_args(argc, argv);
120 rte_exit(EXIT_FAILURE, "Error with PTP initialization\n");
122 The ``main()`` also allocates a mempool to hold the mbufs (Message Buffers)
123 used by the application:
127 mbuf_pool = rte_pktmbuf_pool_create("MBUF_POOL", NUM_MBUFS * nb_ports,
128 MBUF_CACHE_SIZE, 0, RTE_MBUF_DEFAULT_BUF_SIZE, rte_socket_id());
130 Mbufs are the packet buffer structure used by DPDK. They are explained in
131 detail in the "Mbuf Library" section of the *DPDK Programmer's Guide*.
133 The ``main()`` function also initializes all the ports using the user defined
134 ``port_init()`` function with portmask provided by user:
138 for (portid = 0; portid < nb_ports; portid++)
139 if ((ptp_enabled_port_mask & (1 << portid)) != 0) {
141 if (port_init(portid, mbuf_pool) == 0) {
142 ptp_enabled_ports[ptp_enabled_port_nb] = portid;
143 ptp_enabled_port_nb++;
145 rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu8 "\n",
151 Once the initialization is complete, the application is ready to launch a
152 function on an lcore. In this example ``lcore_main()`` is called on a single
159 The ``lcore_main()`` function is explained below.
165 As we saw above the ``main()`` function calls an application function on the
168 The main work of the application is done within the loop:
172 for (portid = 0; portid < ptp_enabled_port_nb; portid++) {
174 portid = ptp_enabled_ports[portid];
175 nb_rx = rte_eth_rx_burst(portid, 0, &m, 1);
177 if (likely(nb_rx == 0))
180 if (m->ol_flags & PKT_RX_IEEE1588_PTP)
181 parse_ptp_frames(portid, m);
186 Packets are received one by one on the RX ports and, if required, PTP response
187 packets are transmitted on the TX ports.
189 If the offload flags in the mbuf indicate that the packet is a PTP packet then
190 the packet is parsed to determine which type:
194 if (m->ol_flags & PKT_RX_IEEE1588_PTP)
195 parse_ptp_frames(portid, m);
198 All packets are freed explicitly using ``rte_pktmbuf_free()``.
200 The forwarding loop can be interrupted and the application closed using
207 The ``parse_ptp_frames()`` function processes PTP packets, implementing slave
208 PTP IEEE1588 L2 functionality.
213 parse_ptp_frames(uint16_t portid, struct rte_mbuf *m) {
214 struct ptp_header *ptp_hdr;
215 struct rte_ether_hdr *eth_hdr;
218 eth_hdr = rte_pktmbuf_mtod(m, struct rte_ether_hdr *);
219 eth_type = rte_be_to_cpu_16(eth_hdr->ether_type);
221 if (eth_type == PTP_PROTOCOL) {
223 ptp_data.portid = portid;
224 ptp_hdr = (struct ptp_header *)(rte_pktmbuf_mtod(m, char *)
225 + sizeof(struct rte_ether_hdr));
227 switch (ptp_hdr->msgtype) {
229 parse_sync(&ptp_data);
232 parse_fup(&ptp_data);
235 parse_drsp(&ptp_data);
236 print_clock_info(&ptp_data);
244 There are 3 types of packets on the RX path which we must parse to create a minimal
245 implementation of the PTP slave client:
249 * DELAY RESPONSE packet.
251 When we parse the *FOLLOW UP* packet we also create and send a *DELAY_REQUEST* packet.
252 Also when we parse the *DELAY RESPONSE* packet, and all conditions are met we adjust the PTP slave clock.