xref: /linux/Documentation/networking/j1939.rst (revision c532de5a67a70f8533d495f8f2aaa9a0491c3ad0)
1.. SPDX-License-Identifier: (GPL-2.0 OR MIT)
2
3===================
4J1939 Documentation
5===================
6
7Overview / What Is J1939
8========================
9
10SAE J1939 defines a higher layer protocol on CAN. It implements a more
11sophisticated addressing scheme and extends the maximum packet size above 8
12bytes. Several derived specifications exist, which differ from the original
13J1939 on the application level, like MilCAN A, NMEA2000, and especially
14ISO-11783 (ISOBUS). This last one specifies the so-called ETP (Extended
15Transport Protocol), which has been included in this implementation. This
16results in a maximum packet size of ((2 ^ 24) - 1) * 7 bytes == 111 MiB.
17
18Specifications used
19-------------------
20
21* SAE J1939-21 : data link layer
22* SAE J1939-81 : network management
23* ISO 11783-6  : Virtual Terminal (Extended Transport Protocol)
24
25.. _j1939-motivation:
26
27Motivation
28==========
29
30Given the fact there's something like SocketCAN with an API similar to BSD
31sockets, we found some reasons to justify a kernel implementation for the
32addressing and transport methods used by J1939.
33
34* **Addressing:** when a process on an ECU communicates via J1939, it should
35  not necessarily know its source address. Although, at least one process per
36  ECU should know the source address. Other processes should be able to reuse
37  that address. This way, address parameters for different processes
38  cooperating for the same ECU, are not duplicated. This way of working is
39  closely related to the UNIX concept, where programs do just one thing and do
40  it well.
41
42* **Dynamic addressing:** Address Claiming in J1939 is time critical.
43  Furthermore, data transport should be handled properly during the address
44  negotiation. Putting this functionality in the kernel eliminates it as a
45  requirement for _every_ user space process that communicates via J1939. This
46  results in a consistent J1939 bus with proper addressing.
47
48* **Transport:** both TP & ETP reuse some PGNs to relay big packets over them.
49  Different processes may thus use the same TP & ETP PGNs without actually
50  knowing it. The individual TP & ETP sessions _must_ be serialized
51  (synchronized) between different processes. The kernel solves this problem
52  properly and eliminates the serialization (synchronization) as a requirement
53  for _every_ user space process that communicates via J1939.
54
55J1939 defines some other features (relaying, gateway, fast packet transport,
56...). In-kernel code for these would not contribute to protocol stability.
57Therefore, these parts are left to user space.
58
59The J1939 sockets operate on CAN network devices (see SocketCAN). Any J1939
60user space library operating on CAN raw sockets will still operate properly.
61Since such a library does not communicate with the in-kernel implementation, care
62must be taken that these two do not interfere. In practice, this means they
63cannot share ECU addresses. A single ECU (or virtual ECU) address is used by
64the library exclusively, or by the in-kernel system exclusively.
65
66J1939 concepts
67==============
68
69PGN
70---
71
72The J1939 protocol uses the 29-bit CAN identifier with the following structure:
73
74  ============  ==============  ====================
75  29 bit CAN-ID
76  --------------------------------------------------
77  Bit positions within the CAN-ID
78  --------------------------------------------------
79  28 ... 26     25 ... 8        7 ... 0
80  ============  ==============  ====================
81  Priority      PGN             SA (Source Address)
82  ============  ==============  ====================
83
84The PGN (Parameter Group Number) is a number to identify a packet. The PGN
85is composed as follows:
86
87  ============  ==============  =================  =================
88  PGN
89  ------------------------------------------------------------------
90  Bit positions within the CAN-ID
91  ------------------------------------------------------------------
92  25            24              23 ... 16          15 ... 8
93  ============  ==============  =================  =================
94  R (Reserved)  DP (Data Page)  PF (PDU Format)    PS (PDU Specific)
95  ============  ==============  =================  =================
96
97In J1939-21 distinction is made between PDU1 format (where PF < 240) and PDU2
98format (where PF >= 240). Furthermore, when using the PDU2 format, the PS-field
99contains a so-called Group Extension, which is part of the PGN. When using PDU2
100format, the Group Extension is set in the PS-field.
101
102  ==============  ========================
103  PDU1 Format (specific) (peer to peer)
104  ----------------------------------------
105  Bit positions within the CAN-ID
106  ----------------------------------------
107  23 ... 16       15 ... 8
108  ==============  ========================
109  00h ... EFh     DA (Destination address)
110  ==============  ========================
111
112  ==============  ========================
113  PDU2 Format (global) (broadcast)
114  ----------------------------------------
115  Bit positions within the CAN-ID
116  ----------------------------------------
117  23 ... 16       15 ... 8
118  ==============  ========================
119  F0h ... FFh     GE (Group Extension)
120  ==============  ========================
121
122On the other hand, when using PDU1 format, the PS-field contains a so-called
123Destination Address, which is _not_ part of the PGN. When communicating a PGN
124from user space to kernel (or vice versa) and PDU1 format is used, the PS-field
125of the PGN shall be set to zero. The Destination Address shall be set
126elsewhere.
127
128Regarding PGN mapping to 29-bit CAN identifier, the Destination Address shall
129be get/set from/to the appropriate bits of the identifier by the kernel.
130
131
132Addressing
133----------
134
135Both static and dynamic addressing methods can be used.
136
137For static addresses, no extra checks are made by the kernel and provided
138addresses are considered right. This responsibility is for the OEM or system
139integrator.
140
141For dynamic addressing, so-called Address Claiming, extra support is foreseen
142in the kernel. In J1939 any ECU is known by its 64-bit NAME. At the moment of
143a successful address claim, the kernel keeps track of both NAME and source
144address being claimed. This serves as a base for filter schemes. By default,
145packets with a destination that is not locally will be rejected.
146
147Mixed mode packets (from a static to a dynamic address or vice versa) are
148allowed. The BSD sockets define separate API calls for getting/setting the
149local & remote address and are applicable for J1939 sockets.
150
151Filtering
152---------
153
154J1939 defines white list filters per socket that a user can set in order to
155receive a subset of the J1939 traffic. Filtering can be based on:
156
157* SA
158* SOURCE_NAME
159* PGN
160
161When multiple filters are in place for a single socket, and a packet comes in
162that matches several of those filters, the packet is only received once for
163that socket.
164
165How to Use J1939
166================
167
168API Calls
169---------
170
171On CAN, you first need to open a socket for communicating over a CAN network.
172To use J1939, ``#include <linux/can/j1939.h>``. From there, ``<linux/can.h>`` will be
173included too. To open a socket, use:
174
175.. code-block:: C
176
177    s = socket(PF_CAN, SOCK_DGRAM, CAN_J1939);
178
179J1939 does use ``SOCK_DGRAM`` sockets. In the J1939 specification, connections are
180mentioned in the context of transport protocol sessions. These still deliver
181packets to the other end (using several CAN packets). ``SOCK_STREAM`` is not
182supported.
183
184After the successful creation of the socket, you would normally use the ``bind(2)``
185and/or ``connect(2)`` system call to bind the socket to a CAN interface. After
186binding and/or connecting the socket, you can ``read(2)`` and ``write(2)`` from/to the
187socket or use ``send(2)``, ``sendto(2)``, ``sendmsg(2)`` and the ``recv*()`` counterpart
188operations on the socket as usual. There are also J1939 specific socket options
189described below.
190
191In order to send data, a ``bind(2)`` must have been successful. ``bind(2)`` assigns a
192local address to a socket.
193
194Different from CAN is that the payload data is just the data that get sends,
195without its header info. The header info is derived from the sockaddr supplied
196to ``bind(2)``, ``connect(2)``, ``sendto(2)`` and ``recvfrom(2)``. A ``write(2)`` with size 4 will
197result in a packet with 4 bytes.
198
199The sockaddr structure has extensions for use with J1939 as specified below:
200
201.. code-block:: C
202
203      struct sockaddr_can {
204         sa_family_t can_family;
205         int         can_ifindex;
206         union {
207            struct {
208               __u64 name;
209                        /* pgn:
210                         * 8 bit: PS in PDU2 case, else 0
211                         * 8 bit: PF
212                         * 1 bit: DP
213                         * 1 bit: reserved
214                         */
215               __u32 pgn;
216               __u8  addr;
217            } j1939;
218         } can_addr;
219      }
220
221``can_family`` & ``can_ifindex`` serve the same purpose as for other SocketCAN sockets.
222
223``can_addr.j1939.pgn`` specifies the PGN (max 0x3ffff). Individual bits are
224specified above.
225
226``can_addr.j1939.name`` contains the 64-bit J1939 NAME.
227
228``can_addr.j1939.addr`` contains the address.
229
230The ``bind(2)`` system call assigns the local address, i.e. the source address when
231sending packages. If a PGN during ``bind(2)`` is set, it's used as a RX filter.
232I.e. only packets with a matching PGN are received. If an ADDR or NAME is set
233it is used as a receive filter, too. It will match the destination NAME or ADDR
234of the incoming packet. The NAME filter will work only if appropriate Address
235Claiming for this name was done on the CAN bus and registered/cached by the
236kernel.
237
238On the other hand ``connect(2)`` assigns the remote address, i.e. the destination
239address. The PGN from ``connect(2)`` is used as the default PGN when sending
240packets. If ADDR or NAME is set it will be used as the default destination ADDR
241or NAME. Further a set ADDR or NAME during ``connect(2)`` is used as a receive
242filter. It will match the source NAME or ADDR of the incoming packet.
243
244Both ``write(2)`` and ``send(2)`` will send a packet with local address from ``bind(2)`` and the
245remote address from ``connect(2)``. Use ``sendto(2)`` to overwrite the destination
246address.
247
248If ``can_addr.j1939.name`` is set (!= 0) the NAME is looked up by the kernel and
249the corresponding ADDR is used. If ``can_addr.j1939.name`` is not set (== 0),
250``can_addr.j1939.addr`` is used.
251
252When creating a socket, reasonable defaults are set. Some options can be
253modified with ``setsockopt(2)`` & ``getsockopt(2)``.
254
255RX path related options:
256
257- ``SO_J1939_FILTER`` - configure array of filters
258- ``SO_J1939_PROMISC`` - disable filters set by ``bind(2)`` and ``connect(2)``
259
260By default no broadcast packets can be send or received. To enable sending or
261receiving broadcast packets use the socket option ``SO_BROADCAST``:
262
263.. code-block:: C
264
265     int value = 1;
266     setsockopt(sock, SOL_SOCKET, SO_BROADCAST, &value, sizeof(value));
267
268The following diagram illustrates the RX path:
269
270.. code::
271
272                    +--------------------+
273                    |  incoming packet   |
274                    +--------------------+
275                              |
276                              V
277                    +--------------------+
278                    | SO_J1939_PROMISC?  |
279                    +--------------------+
280                             |  |
281                         no  |  | yes
282                             |  |
283                   .---------'  `---------.
284                   |                      |
285     +---------------------------+        |
286     | bind() + connect() +      |        |
287     | SOCK_BROADCAST filter     |        |
288     +---------------------------+        |
289                   |                      |
290                   |<---------------------'
291                   V
292     +---------------------------+
293     |      SO_J1939_FILTER      |
294     +---------------------------+
295                   |
296                   V
297     +---------------------------+
298     |        socket recv()      |
299     +---------------------------+
300
301TX path related options:
302``SO_J1939_SEND_PRIO`` - change default send priority for the socket
303
304Message Flags during send() and Related System Calls
305^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
306
307``send(2)``, ``sendto(2)`` and ``sendmsg(2)`` take a 'flags' argument. Currently
308supported flags are:
309
310* ``MSG_DONTWAIT``, i.e. non-blocking operation.
311
312recvmsg(2)
313^^^^^^^^^^
314
315In most cases ``recvmsg(2)`` is needed if you want to extract more information than
316``recvfrom(2)`` can provide. For example package priority and timestamp. The
317Destination Address, name and packet priority (if applicable) are attached to
318the msghdr in the ``recvmsg(2)`` call. They can be extracted using ``cmsg(3)`` macros,
319with ``cmsg_level == SOL_J1939 && cmsg_type == SCM_J1939_DEST_ADDR``,
320``SCM_J1939_DEST_NAME`` or ``SCM_J1939_PRIO``. The returned data is a ``uint8_t`` for
321``priority`` and ``dst_addr``, and ``uint64_t`` for ``dst_name``.
322
323.. code-block:: C
324
325	uint8_t priority, dst_addr;
326	uint64_t dst_name;
327
328	for (cmsg = CMSG_FIRSTHDR(&msg); cmsg; cmsg = CMSG_NXTHDR(&msg, cmsg)) {
329		switch (cmsg->cmsg_level) {
330		case SOL_CAN_J1939:
331			if (cmsg->cmsg_type == SCM_J1939_DEST_ADDR)
332				dst_addr = *CMSG_DATA(cmsg);
333			else if (cmsg->cmsg_type == SCM_J1939_DEST_NAME)
334				memcpy(&dst_name, CMSG_DATA(cmsg), cmsg->cmsg_len - CMSG_LEN(0));
335			else if (cmsg->cmsg_type == SCM_J1939_PRIO)
336				priority = *CMSG_DATA(cmsg);
337			break;
338		}
339	}
340
341Dynamic Addressing
342------------------
343
344Distinction has to be made between using the claimed address and doing an
345address claim. To use an already claimed address, one has to fill in the
346``j1939.name`` member and provide it to ``bind(2)``. If the name had claimed an address
347earlier, all further messages being sent will use that address. And the
348``j1939.addr`` member will be ignored.
349
350An exception on this is PGN 0x0ee00. This is the "Address Claim/Cannot Claim
351Address" message and the kernel will use the ``j1939.addr`` member for that PGN if
352necessary.
353
354To claim an address following code example can be used:
355
356.. code-block:: C
357
358	struct sockaddr_can baddr = {
359		.can_family = AF_CAN,
360		.can_addr.j1939 = {
361			.name = name,
362			.addr = J1939_IDLE_ADDR,
363			.pgn = J1939_NO_PGN,	/* to disable bind() rx filter for PGN */
364		},
365		.can_ifindex = if_nametoindex("can0"),
366	};
367
368	bind(sock, (struct sockaddr *)&baddr, sizeof(baddr));
369
370	/* for Address Claiming broadcast must be allowed */
371	int value = 1;
372	setsockopt(sock, SOL_SOCKET, SO_BROADCAST, &value, sizeof(value));
373
374	/* configured advanced RX filter with PGN needed for Address Claiming */
375	const struct j1939_filter filt[] = {
376		{
377			.pgn = J1939_PGN_ADDRESS_CLAIMED,
378			.pgn_mask = J1939_PGN_PDU1_MAX,
379		}, {
380			.pgn = J1939_PGN_REQUEST,
381			.pgn_mask = J1939_PGN_PDU1_MAX,
382		}, {
383			.pgn = J1939_PGN_ADDRESS_COMMANDED,
384			.pgn_mask = J1939_PGN_MAX,
385		},
386	};
387
388	setsockopt(sock, SOL_CAN_J1939, SO_J1939_FILTER, &filt, sizeof(filt));
389
390	uint64_t dat = htole64(name);
391	const struct sockaddr_can saddr = {
392		.can_family = AF_CAN,
393		.can_addr.j1939 = {
394			.pgn = J1939_PGN_ADDRESS_CLAIMED,
395			.addr = J1939_NO_ADDR,
396		},
397	};
398
399	/* Afterwards do a sendto(2) with data set to the NAME (Little Endian). If the
400	 * NAME provided, does not match the j1939.name provided to bind(2), EPROTO
401	 * will be returned.
402	 */
403	sendto(sock, dat, sizeof(dat), 0, (const struct sockaddr *)&saddr, sizeof(saddr));
404
405If no-one else contests the address claim within 250ms after transmission, the
406kernel marks the NAME-SA assignment as valid. The valid assignment will be kept
407among other valid NAME-SA assignments. From that point, any socket bound to the
408NAME can send packets.
409
410If another ECU claims the address, the kernel will mark the NAME-SA expired.
411No socket bound to the NAME can send packets (other than address claims). To
412claim another address, some socket bound to NAME, must ``bind(2)`` again, but with
413only ``j1939.addr`` changed to the new SA, and must then send a valid address claim
414packet. This restarts the state machine in the kernel (and any other
415participant on the bus) for this NAME.
416
417``can-utils`` also include the ``j1939acd`` tool, so it can be used as code example or as
418default Address Claiming daemon.
419
420Send Examples
421-------------
422
423Static Addressing
424^^^^^^^^^^^^^^^^^
425
426This example will send a PGN (0x12300) from SA 0x20 to DA 0x30.
427
428Bind:
429
430.. code-block:: C
431
432	struct sockaddr_can baddr = {
433		.can_family = AF_CAN,
434		.can_addr.j1939 = {
435			.name = J1939_NO_NAME,
436			.addr = 0x20,
437			.pgn = J1939_NO_PGN,
438		},
439		.can_ifindex = if_nametoindex("can0"),
440	};
441
442	bind(sock, (struct sockaddr *)&baddr, sizeof(baddr));
443
444Now, the socket 'sock' is bound to the SA 0x20. Since no ``connect(2)`` was called,
445at this point we can use only ``sendto(2)`` or ``sendmsg(2)``.
446
447Send:
448
449.. code-block:: C
450
451	const struct sockaddr_can saddr = {
452		.can_family = AF_CAN,
453		.can_addr.j1939 = {
454			.name = J1939_NO_NAME;
455			.addr = 0x30,
456			.pgn = 0x12300,
457		},
458	};
459
460	sendto(sock, dat, sizeof(dat), 0, (const struct sockaddr *)&saddr, sizeof(saddr));
461