| 1 |
.. SPDX-License-Identifier: GFDL-1.1-no-invariants-or-later |
| 2 |
|
| 3 |
.. _cec_pin_error_inj: |
| 4 |
|
| 5 |
CEC Pin Framework Error Injection |
| 6 |
================================= |
| 7 |
|
| 8 |
The CEC Pin Framework is a core CEC framework for CEC hardware that only |
| 9 |
has low-level support for the CEC bus. Most hardware today will have |
| 10 |
high-level CEC support where the hardware deals with driving the CEC bus, |
| 11 |
but some older devices aren't that fancy. However, this framework also |
| 12 |
allows you to connect the CEC pin to a GPIO on e.g. a Raspberry Pi and |
| 13 |
you have now made a CEC adapter. |
| 14 |
|
| 15 |
What makes doing this so interesting is that since we have full control |
| 16 |
over the bus it is easy to support error injection. This is ideal to |
| 17 |
test how well CEC adapters can handle error conditions. |
| 18 |
|
| 19 |
Currently only the cec-gpio driver (when the CEC line is directly |
| 20 |
connected to a pull-up GPIO line) and the AllWinner A10/A20 drm driver |
| 21 |
support this framework. |
| 22 |
|
| 23 |
If ``CONFIG_CEC_PIN_ERROR_INJ`` is enabled, then error injection is available |
| 24 |
through debugfs. Specifically, in ``/sys/kernel/debug/cec/cecX/`` there is |
| 25 |
now an ``error-inj`` file. |
| 26 |
|
| 27 |
.. note:: |
| 28 |
|
| 29 |
The error injection commands are not a stable ABI and may change in the |
| 30 |
future. |
| 31 |
|
| 32 |
With ``cat error-inj`` you can see both the possible commands and the current |
| 33 |
error injection status:: |
| 34 |
|
| 35 |
$ cat /sys/kernel/debug/cec/cec0/error-inj |
| 36 |
# Clear error injections: |
| 37 |
# clear clear all rx and tx error injections |
| 38 |
# rx-clear clear all rx error injections |
| 39 |
# tx-clear clear all tx error injections |
| 40 |
# <op> clear clear all rx and tx error injections for <op> |
| 41 |
# <op> rx-clear clear all rx error injections for <op> |
| 42 |
# <op> tx-clear clear all tx error injections for <op> |
| 43 |
# |
| 44 |
# RX error injection settings: |
| 45 |
# rx-no-low-drive do not generate low-drive pulses |
| 46 |
# |
| 47 |
# RX error injection: |
| 48 |
# <op>[,<mode>] rx-nack NACK the message instead of sending an ACK |
| 49 |
# <op>[,<mode>] rx-low-drive <bit> force a low-drive condition at this bit position |
| 50 |
# <op>[,<mode>] rx-add-byte add a spurious byte to the received CEC message |
| 51 |
# <op>[,<mode>] rx-remove-byte remove the last byte from the received CEC message |
| 52 |
# any[,<mode>] rx-arb-lost [<poll>] generate a POLL message to trigger an arbitration lost |
| 53 |
# |
| 54 |
# TX error injection settings: |
| 55 |
# tx-ignore-nack-until-eom ignore early NACKs until EOM |
| 56 |
# tx-custom-low-usecs <usecs> define the 'low' time for the custom pulse |
| 57 |
# tx-custom-high-usecs <usecs> define the 'high' time for the custom pulse |
| 58 |
# tx-custom-pulse transmit the custom pulse once the bus is idle |
| 59 |
# tx-glitch-low-usecs <usecs> define the 'low' time for the glitch pulse |
| 60 |
# tx-glitch-high-usecs <usecs> define the 'high' time for the glitch pulse |
| 61 |
# tx-glitch-falling-edge send the glitch pulse after every falling edge |
| 62 |
# tx-glitch-rising-edge send the glitch pulse after every rising edge |
| 63 |
# |
| 64 |
# TX error injection: |
| 65 |
# <op>[,<mode>] tx-no-eom don't set the EOM bit |
| 66 |
# <op>[,<mode>] tx-early-eom set the EOM bit one byte too soon |
| 67 |
# <op>[,<mode>] tx-add-bytes <num> append <num> (1-255) spurious bytes to the message |
| 68 |
# <op>[,<mode>] tx-remove-byte drop the last byte from the message |
| 69 |
# <op>[,<mode>] tx-short-bit <bit> make this bit shorter than allowed |
| 70 |
# <op>[,<mode>] tx-long-bit <bit> make this bit longer than allowed |
| 71 |
# <op>[,<mode>] tx-custom-bit <bit> send the custom pulse instead of this bit |
| 72 |
# <op>[,<mode>] tx-short-start send a start pulse that's too short |
| 73 |
# <op>[,<mode>] tx-long-start send a start pulse that's too long |
| 74 |
# <op>[,<mode>] tx-custom-start send the custom pulse instead of the start pulse |
| 75 |
# <op>[,<mode>] tx-last-bit <bit> stop sending after this bit |
| 76 |
# <op>[,<mode>] tx-low-drive <bit> force a low-drive condition at this bit position |
| 77 |
# |
| 78 |
# <op> CEC message opcode (0-255) or 'any' |
| 79 |
# <mode> 'once' (default), 'always', 'toggle' or 'off' |
| 80 |
# <bit> CEC message bit (0-159) |
| 81 |
# 10 bits per 'byte': bits 0-7: data, bit 8: EOM, bit 9: ACK |
| 82 |
# <poll> CEC poll message used to test arbitration lost (0x00-0xff, default 0x0f) |
| 83 |
# <usecs> microseconds (0-10000000, default 1000) |
| 84 |
|
| 85 |
clear |
| 86 |
|
| 87 |
You can write error injection commands to ``error-inj`` using |
| 88 |
``echo 'cmd' >error-inj`` or ``cat cmd.txt >error-inj``. The ``cat error-inj`` |
| 89 |
output contains the current error commands. You can save the output to a file |
| 90 |
and use it as an input to ``error-inj`` later. |
| 91 |
|
| 92 |
Basic Syntax |
| 93 |
------------ |
| 94 |
|
| 95 |
Leading spaces/tabs are ignored. If the next character is a ``#`` or the end |
| 96 |
of the line was reached, then the whole line is ignored. Otherwise a command |
| 97 |
is expected. |
| 98 |
|
| 99 |
The error injection commands fall in two main groups: those relating to |
| 100 |
receiving CEC messages and those relating to transmitting CEC messages. In |
| 101 |
addition, there are commands to clear existing error injection commands and |
| 102 |
to create custom pulses on the CEC bus. |
| 103 |
|
| 104 |
Most error injection commands can be executed for specific CEC opcodes or for |
| 105 |
all opcodes (``any``). Each command also has a 'mode' which can be ``off`` |
| 106 |
(can be used to turn off an existing error injection command), ``once`` |
| 107 |
(the default) which will trigger the error injection only once for the next |
| 108 |
received or transmitted message, ``always`` to always trigger the error |
| 109 |
injection and ``toggle`` to toggle the error injection on or off for every |
| 110 |
transmit or receive. |
| 111 |
|
| 112 |
So '``any rx-nack``' will NACK the next received CEC message, |
| 113 |
'``any,always rx-nack``' will NACK all received CEC messages and |
| 114 |
'``0x82,toggle rx-nack``' will only NACK if an Active Source message was |
| 115 |
received and do that only for every other received message. |
| 116 |
|
| 117 |
After an error was injected with mode ``once`` the error injection command |
| 118 |
is cleared automatically, so ``once`` is a one-time deal. |
| 119 |
|
| 120 |
All combinations of ``<op>`` and error injection commands can co-exist. So |
| 121 |
this is fine:: |
| 122 |
|
| 123 |
0x9e tx-add-bytes 1 |
| 124 |
0x9e tx-early-eom |
| 125 |
0x9f tx-add-bytes 2 |
| 126 |
any rx-nack |
| 127 |
|
| 128 |
All four error injection commands will be active simultaneously. |
| 129 |
|
| 130 |
However, if the same ``<op>`` and command combination is specified, |
| 131 |
but with different arguments:: |
| 132 |
|
| 133 |
0x9e tx-add-bytes 1 |
| 134 |
0x9e tx-add-bytes 2 |
| 135 |
|
| 136 |
Then the second will overwrite the first. |
| 137 |
|
| 138 |
Clear Error Injections |
| 139 |
---------------------- |
| 140 |
|
| 141 |
``clear`` |
| 142 |
Clear all error injections. |
| 143 |
|
| 144 |
``rx-clear`` |
| 145 |
Clear all receive error injections |
| 146 |
|
| 147 |
``tx-clear`` |
| 148 |
Clear all transmit error injections |
| 149 |
|
| 150 |
``<op> clear`` |
| 151 |
Clear all error injections for the given opcode. |
| 152 |
|
| 153 |
``<op> rx-clear`` |
| 154 |
Clear all receive error injections for the given opcode. |
| 155 |
|
| 156 |
``<op> tx-clear`` |
| 157 |
Clear all transmit error injections for the given opcode. |
| 158 |
|
| 159 |
Receive Messages |
| 160 |
---------------- |
| 161 |
|
| 162 |
``<op>[,<mode>] rx-nack`` |
| 163 |
NACK broadcast messages and messages directed to this CEC adapter. |
| 164 |
Every byte of the message will be NACKed in case the transmitter |
| 165 |
keeps transmitting after the first byte was NACKed. |
| 166 |
|
| 167 |
``<op>[,<mode>] rx-low-drive <bit>`` |
| 168 |
Force a Low Drive condition at this bit position. If <op> specifies |
| 169 |
a specific CEC opcode then the bit position must be at least 18, |
| 170 |
otherwise the opcode hasn't been received yet. This tests if the |
| 171 |
transmitter can handle the Low Drive condition correctly and reports |
| 172 |
the error correctly. Note that a Low Drive in the first 4 bits can also |
| 173 |
be interpreted as an Arbitration Lost condition by the transmitter. |
| 174 |
This is implementation dependent. |
| 175 |
|
| 176 |
``<op>[,<mode>] rx-add-byte`` |
| 177 |
Add a spurious 0x55 byte to the received CEC message, provided |
| 178 |
the message was 15 bytes long or less. This is useful to test |
| 179 |
the high-level protocol since spurious bytes should be ignored. |
| 180 |
|
| 181 |
``<op>[,<mode>] rx-remove-byte`` |
| 182 |
Remove the last byte from the received CEC message, provided it |
| 183 |
was at least 2 bytes long. This is useful to test the high-level |
| 184 |
protocol since messages that are too short should be ignored. |
| 185 |
|
| 186 |
``<op>[,<mode>] rx-arb-lost <poll>`` |
| 187 |
Generate a POLL message to trigger an Arbitration Lost condition. |
| 188 |
This command is only allowed for ``<op>`` values of ``next`` or ``all``. |
| 189 |
As soon as a start bit has been received the CEC adapter will switch |
| 190 |
to transmit mode and it will transmit a POLL message. By default this is |
| 191 |
0x0f, but it can also be specified explicitly via the ``<poll>`` argument. |
| 192 |
|
| 193 |
This command can be used to test the Arbitration Lost condition in |
| 194 |
the remote CEC transmitter. Arbitration happens when two CEC adapters |
| 195 |
start sending a message at the same time. In that case the initiator |
| 196 |
with the most leading zeroes wins and the other transmitter has to |
| 197 |
stop transmitting ('Arbitration Lost'). This is very hard to test, |
| 198 |
except by using this error injection command. |
| 199 |
|
| 200 |
This does not work if the remote CEC transmitter has logical address |
| 201 |
0 ('TV') since that will always win. |
| 202 |
|
| 203 |
``rx-no-low-drive`` |
| 204 |
The receiver will ignore situations that would normally generate a |
| 205 |
Low Drive pulse (3.6 ms). This is typically done if a spurious pulse is |
| 206 |
detected when receiving a message, and it indicates to the transmitter that |
| 207 |
the message has to be retransmitted since the receiver got confused. |
| 208 |
Disabling this is useful to test how other CEC devices handle glitches |
| 209 |
by ensuring we will not be the one that generates a Low Drive. |
| 210 |
|
| 211 |
Transmit Messages |
| 212 |
----------------- |
| 213 |
|
| 214 |
``tx-ignore-nack-until-eom`` |
| 215 |
This setting changes the behavior of transmitting CEC messages. Normally |
| 216 |
as soon as the receiver NACKs a byte the transmit will stop, but the |
| 217 |
specification also allows that the full message is transmitted and only |
| 218 |
at the end will the transmitter look at the ACK bit. This is not |
| 219 |
recommended behavior since there is no point in keeping the CEC bus busy |
| 220 |
for longer than is strictly needed. Especially given how slow the bus is. |
| 221 |
|
| 222 |
This setting can be used to test how well a receiver deals with |
| 223 |
transmitters that ignore NACKs until the very end of the message. |
| 224 |
|
| 225 |
``<op>[,<mode>] tx-no-eom`` |
| 226 |
Don't set the EOM bit. Normally the last byte of the message has the EOM |
| 227 |
(End-Of-Message) bit set. With this command the transmit will just stop |
| 228 |
without ever sending an EOM. This can be used to test how a receiver |
| 229 |
handles this case. Normally receivers have a time-out after which |
| 230 |
they will go back to the Idle state. |
| 231 |
|
| 232 |
``<op>[,<mode>] tx-early-eom`` |
| 233 |
Set the EOM bit one byte too soon. This obviously only works for messages |
| 234 |
of two bytes or more. The EOM bit will be set for the second-to-last byte |
| 235 |
and not for the final byte. The receiver should ignore the last byte in |
| 236 |
this case. Since the resulting message is likely to be too short for this |
| 237 |
same reason the whole message is typically ignored. The receiver should be |
| 238 |
in Idle state after the last byte was transmitted. |
| 239 |
|
| 240 |
``<op>[,<mode>] tx-add-bytes <num>`` |
| 241 |
Append ``<num>`` (1-255) spurious bytes to the message. The extra bytes |
| 242 |
have the value of the byte position in the message. So if you transmit a |
| 243 |
two byte message (e.g. a Get CEC Version message) and add 2 bytes, then |
| 244 |
the full message received by the remote CEC adapter is |
| 245 |
``0x40 0x9f 0x02 0x03``. |
| 246 |
|
| 247 |
This command can be used to test buffer overflows in the receiver. E.g. |
| 248 |
what does it do when it receives more than the maximum message size of 16 |
| 249 |
bytes. |
| 250 |
|
| 251 |
``<op>[,<mode>] tx-remove-byte`` |
| 252 |
Drop the last byte from the message, provided the message is at least |
| 253 |
two bytes long. The receiver should ignore messages that are too short. |
| 254 |
|
| 255 |
``<op>[,<mode>] tx-short-bit <bit>`` |
| 256 |
Make this bit period shorter than allowed. The bit position cannot be |
| 257 |
an Ack bit. If <op> specifies a specific CEC opcode then the bit position |
| 258 |
must be at least 18, otherwise the opcode hasn't been received yet. |
| 259 |
Normally the period of a data bit is between 2.05 and 2.75 milliseconds. |
| 260 |
With this command the period of this bit is 1.8 milliseconds, this is |
| 261 |
done by reducing the time the CEC bus is high. This bit period is less |
| 262 |
than is allowed and the receiver should respond with a Low Drive |
| 263 |
condition. |
| 264 |
|
| 265 |
This command is ignored for 0 bits in bit positions 0 to 3. This is |
| 266 |
because the receiver also looks for an Arbitration Lost condition in |
| 267 |
those first four bits and it is undefined what will happen if it |
| 268 |
sees a too-short 0 bit. |
| 269 |
|
| 270 |
``<op>[,<mode>] tx-long-bit <bit>`` |
| 271 |
Make this bit period longer than is valid. The bit position cannot be |
| 272 |
an Ack bit. If <op> specifies a specific CEC opcode then the bit position |
| 273 |
must be at least 18, otherwise the opcode hasn't been received yet. |
| 274 |
Normally the period of a data bit is between 2.05 and 2.75 milliseconds. |
| 275 |
With this command the period of this bit is 2.9 milliseconds, this is |
| 276 |
done by increasing the time the CEC bus is high. |
| 277 |
|
| 278 |
Even though this bit period is longer than is valid it is undefined what |
| 279 |
a receiver will do. It might just accept it, or it might time out and |
| 280 |
return to Idle state. Unfortunately the CEC specification is silent about |
| 281 |
this. |
| 282 |
|
| 283 |
This command is ignored for 0 bits in bit positions 0 to 3. This is |
| 284 |
because the receiver also looks for an Arbitration Lost condition in |
| 285 |
those first four bits and it is undefined what will happen if it |
| 286 |
sees a too-long 0 bit. |
| 287 |
|
| 288 |
``<op>[,<mode>] tx-short-start`` |
| 289 |
Make this start bit period shorter than allowed. Normally the period of |
| 290 |
a start bit is between 4.3 and 4.7 milliseconds. With this command the |
| 291 |
period of the start bit is 4.1 milliseconds, this is done by reducing |
| 292 |
the time the CEC bus is high. This start bit period is less than is |
| 293 |
allowed and the receiver should return to Idle state when this is detected. |
| 294 |
|
| 295 |
``<op>[,<mode>] tx-long-start`` |
| 296 |
Make this start bit period longer than is valid. Normally the period of |
| 297 |
a start bit is between 4.3 and 4.7 milliseconds. With this command the |
| 298 |
period of the start bit is 5 milliseconds, this is done by increasing |
| 299 |
the time the CEC bus is high. This start bit period is more than is |
| 300 |
valid and the receiver should return to Idle state when this is detected. |
| 301 |
|
| 302 |
Even though this start bit period is longer than is valid it is undefined |
| 303 |
what a receiver will do. It might just accept it, or it might time out and |
| 304 |
return to Idle state. Unfortunately the CEC specification is silent about |
| 305 |
this. |
| 306 |
|
| 307 |
``<op>[,<mode>] tx-last-bit <bit>`` |
| 308 |
Just stop transmitting after this bit. If <op> specifies a specific CEC |
| 309 |
opcode then the bit position must be at least 18, otherwise the opcode |
| 310 |
hasn't been received yet. This command can be used to test how the receiver |
| 311 |
reacts when a message just suddenly stops. It should time out and go back |
| 312 |
to Idle state. |
| 313 |
|
| 314 |
``<op>[,<mode>] tx-low-drive <bit>`` |
| 315 |
Force a Low Drive condition at this bit position. If <op> specifies a |
| 316 |
specific CEC opcode then the bit position must be at least 18, otherwise |
| 317 |
the opcode hasn't been received yet. This can be used to test how the |
| 318 |
receiver handles Low Drive conditions. Note that if this happens at bit |
| 319 |
positions 0-3 the receiver can interpret this as an Arbitration Lost |
| 320 |
condition. This is implementation dependent. |
| 321 |
|
| 322 |
Custom Pulses |
| 323 |
------------- |
| 324 |
|
| 325 |
``tx-custom-low-usecs <usecs>`` |
| 326 |
This defines the duration in microseconds that the custom pulse pulls |
| 327 |
the CEC line low. The default is 1000 microseconds. |
| 328 |
|
| 329 |
``tx-custom-high-usecs <usecs>`` |
| 330 |
This defines the duration in microseconds that the custom pulse keeps the |
| 331 |
CEC line high (unless another CEC adapter pulls it low in that time). |
| 332 |
The default is 1000 microseconds. The total period of the custom pulse is |
| 333 |
``tx-custom-low-usecs + tx-custom-high-usecs``. |
| 334 |
|
| 335 |
``<op>[,<mode>] tx-custom-bit <bit>`` |
| 336 |
Send the custom bit instead of a regular data bit. The bit position cannot |
| 337 |
be an Ack bit. If <op> specifies a specific CEC opcode then the bit |
| 338 |
position must be at least 18, otherwise the opcode hasn't been received yet. |
| 339 |
|
| 340 |
``<op>[,<mode>] tx-custom-start`` |
| 341 |
Send the custom bit instead of a regular start bit. |
| 342 |
|
| 343 |
``tx-custom-pulse`` |
| 344 |
Transmit a single custom pulse as soon as the CEC bus is idle. |
| 345 |
|
| 346 |
Glitch Pulses |
| 347 |
------------- |
| 348 |
|
| 349 |
This emulates what happens if the signal on the CEC line is seeing spurious |
| 350 |
pulses. Typically this happens after the falling or rising edge where there |
| 351 |
is a short voltage fluctuation that, if the CEC hardware doesn't do |
| 352 |
deglitching, can be seen as a spurious pulse and can cause a Low Drive |
| 353 |
condition or corrupt data. |
| 354 |
|
| 355 |
``tx-glitch-low-usecs <usecs>`` |
| 356 |
This defines the duration in microseconds that the glitch pulse pulls |
| 357 |
the CEC line low. The default is 1 microsecond. The range is 0-100 |
| 358 |
microseconds. If 0, then no glitch pulse will be generated. |
| 359 |
|
| 360 |
``tx-glitch-high-usecs <usecs>`` |
| 361 |
This defines the duration in microseconds that the glitch pulse keeps the |
| 362 |
CEC line high (unless another CEC adapter pulls it low in that time). |
| 363 |
The default is 1 microseconds. The range is 0-100 microseconds. If 0, then |
| 364 |
no glitch pulse will be generated.The total period of the glitch pulse is |
| 365 |
``tx-custom-low-usecs + tx-custom-high-usecs``. |
| 366 |
|
| 367 |
``tx-glitch-falling-edge`` |
| 368 |
Send the glitch pulse right after the falling edge. |
| 369 |
|
| 370 |
``tx-glitch-rising-edge`` |
| 371 |
Send the glitch pulse right after the rising edge. |