1 Linux SocketCAN driver for ELM327
2 ==================================
7 Max Staudt <max-linux@enpas.org>
14 CAN adapters are expensive, few, and far between.
15 ELM327 interfaces are cheap and plentiful.
17 This driver aims to lower the initial cost for hackers interested in
18 working with CAN buses.
25 This driver is an effort to turn abundant ELM327 based OBD interfaces
26 into full-fledged (as far as possible) CAN interfaces.
28 Since the ELM327 was never meant to be a stand-alone CAN controller,
29 the driver has to switch between its modes as quickly as possible in
30 order to approximate full-duplex operation.
32 As such, elmcan is a best-effort driver. However, this is more than
33 enough to implement simple request-response protocols (such as OBD II),
34 and to monitor broadcast messages on a bus (such as in a vehicle).
36 Most ELM327s come as nondescript serial devices, attached via USB or
37 Bluetooth. The driver cannot recognize them by itself, and as such it
38 is up to the user to attach it in form of a TTY line discipline
39 (similar to PPP, SLIP, slcan, ...).
41 This driver is meant for ELM327 versions 1.4b and up, see below for
42 known limitations in older controllers and clones.
49 This requires Linux 4.11 (for 431af779256c), and has been tested on 4.19.
56 The official data sheets can be found at ELM electronics' home page:
58 https://www.elmelectronics.com/
62 How to check the controller version
63 ------------------------------------
65 Use a terminal program to attach to the controller.
67 After issuing the "``AT WS``" command, the controller will respond with
79 How to attach the line discipline
80 ----------------------------------
82 Every ELM327 chip is factory programmed to operate at a serial setting
83 of 38400 baud/s, 8 data bits, no parity, 1 stopbit.
85 The line discipline can be attached on a command prompt as follows::
93 --iflag -ICRNL,INLCR,-IXOFF \
97 To change the ELM327's serial settings, please refer to its data
98 sheet. This needs to be done before attaching the line discipline.
105 - ``accept_flaky_uart`` - Try to live with a bad controller or UART line
107 Some adapters and/or their connection are unreliable. Enable this
108 option to try and work around the situation. This is a best-effort
109 workaround, so undefined behavior will likely occur sooner or later.
113 module/kernel parameter: accept_flaky_uart=[0|1]
117 Known limitations of the controller
118 ------------------------------------
120 - Clone "v1.5" devices
122 Sending RTR frames is not supported and will be dropped silently.
124 Receiving RTR with DLC 8 will appear to be a regular frame with
125 the last received frame's DLC and payload.
127 "``AT CSM``" not supported, thus no ACK-ing frames while listening:
128 "``AT MA``" will always be silent. However, immediately after
129 sending a frame, the ELM327 will be in "receive reply" mode, in
130 which it *does* ACK any received frames. Once the bus goes silent
131 or an error occurs (such as BUFFER FULL), the ELM327 will end reply
132 reception mode on its own and elmcan will fall back to "``AT MA``"
133 in order to keep monitoring the bus.
138 No automatic full duplex operation is supported. The driver will
139 switch between input/output mode as quickly as possible.
141 The length of outgoing RTR frames cannot be set. In fact, some
142 clones (tested with one identifying as "``v1.5``") are unable to
143 send RTR frames at all.
145 We don't have a way to get real-time notifications on CAN errors.
146 While there is a command (``AT CS``) to retrieve some basic stats,
147 we don't poll it as it would force us to interrupt reception mode.
150 - Versions prior to 1.4b
152 These versions do not send CAN ACKs when in monitoring mode (AT MA).
153 However, they do send ACKs while waiting for a reply immediately
154 after sending a frame. The driver maximizes this time to make the
155 controller as useful as possible.
157 Starting with version 1.4b, the ELM327 supports the "``AT CSM``"
158 command, and the "listen-only" CAN option will take effect.
161 - Versions prior to 1.4
163 These chips do not support the "``AT PB``" command, and thus cannot
164 change bitrate or SFF/EFF mode on-the-fly. This will have to be
165 programmed by the user before attaching the line discipline. See the
166 data sheet for details.
169 - Versions prior to 1.3
171 These chips cannot be used at all with elmcan. They do not support
172 the "``AT D1``", which is necessary to avoid parsing conflicts on
173 incoming data, as well as distinction of RTR frame lengths.
175 Specifically, this allows for easy distinction of SFF and EFF
176 frames, and to check whether frames are complete. While it is possible
177 to deduce the type and length from the length of the line the ELM327
178 sends us, this method fails when the ELM327's UART output buffer
179 overruns. It may abort sending in the middle of the line, which will
180 then be mistaken for something else.
184 Known limitations of the driver
185 --------------------------------
189 ELM327 can only set CAN bitrates that are of the form 500000/n, where
190 n is an integer divisor.
191 However there is an exception: With a separate flag, it may set the
192 speed to be 8/7 of the speed indicated by the divisor.
193 This mode is not currently implemented.
195 - No evaluation of command responses.
197 The ELM327 will reply with OK when a command is understood, and with ?
198 when it is not. The driver does not currently check this, and simply
199 assumes that the chip understands every command.
200 The driver is built such that functionality degrades gracefully
201 nevertheless. See the section on known limitations of the controller.
203 - No use of hardware CAN ID filtering
205 An ELM327's UART sending buffer will easily overflow on heavy CAN bus
206 load, resulting in the "``BUFFER FULL``" message. Using the hardware
207 filters available through "``AT CF xxx``" and "``AT CM xxx``" would be
208 helpful here, however SocketCAN does not currently provide a facility
209 to make use of such hardware features.
213 Communication example
214 ----------------------
216 This is a short and incomplete introduction on how to talk to an ELM327.
219 The ELM327 has two modes:
224 In command mode, it expects one command per line, terminated by CR.
225 By default, the prompt is a "``>``", after which a command can be
232 The init script in the driver switches off several configuration options
233 that are only meaningful in the original OBD scenario the chip is meant
234 for, and are actually a hindrance for elmcan.
237 When a command is not recognized, such as by an older version of the
238 ELM327, a question mark is printed as a response instead of OK::
244 At present, elmcan does not evaluate this response and silently assumes
245 that all commands are recognized. It is structured such that it will
246 degrade gracefully when a command is unknown. See the sections above on
247 known limitations for details.
250 When a CAN frame is to be sent, the target address is configured, after
251 which the frame is sent as a command that consists of the data's hex
260 The above interaction sends the frame "``DE AD BE EF 12 34 56 78``" with
261 the 11 bit CAN ID ``0x123``.
262 For this to function, the controller must be configured for 11 bit CAN
263 ID sending mode (using "``AT PB``", see code or datasheet).
266 Once a frame has been sent and wait-for-reply mode is on (``ATR1``,
267 configured on ``listen-only=off``), or when the reply timeout expires and
268 the driver sets the controller into monitoring mode (``ATMA``), the ELM327
269 will send one line for each received CAN frame, consisting of CAN ID,
272 123 8 DEADBEEF12345678
274 For 29 bit CAN frames, the address format is slightly different, which
275 elmcan uses to tell the two apart::
277 12 34 56 78 8 DEADBEEF12345678
279 The ELM327 will receive both 11 and 29 bit frames - the current CAN
280 config (``ATPB``) does not matter.
283 If the ELM327's internal UART sending buffer runs full, it will abort
284 the monitoring mode, print "BUFFER FULL" and drop back into command
285 mode. Note that in this case, unlike with other error messages, the
286 error message may appear on the same line as the last (usually
287 incomplete) data frame::
289 12 34 56 78 8 DEADBEEF123 BUFFER FULL
293 Rationale behind the chosen configuration
294 ------------------------------------------
299 We need this to be able to get a prompt reliably.
304 We need this to distinguish 11/29 bit CAN addresses received.
307 We can usually do this using the line length (odd/even),
308 but this fails if the line is not transmitted fully to
309 the host (BUFFER FULL).
314 We need this to tell the "length" of RTR frames.
318 A note on CAN bus termination
319 ------------------------------
321 Your adapter may have resistors soldered in which are meant to terminate
322 the bus. This is correct when it is plugged into a OBD-II socket, but
323 not helpful when trying to tap into the middle of an existing CAN bus.
325 If communications don't work with the adapter connected, check for the
326 termination resistors on its PCB and try removing them.
330 To Do list for future development
331 ----------------------------------
333 - DMA capable rx/tx buffers
334 (is this relevant for this driver?)
336 - flushing of ``tx_work`` is too late in ``ldisc_close()``
337 (is this still the case?)