diff options
| author | Steven Rostedt <rostedt@goodmis.org> | 2026-09-02 09:55:01 -0400 |
|---|---|---|
| committer | Steven Rostedt <rostedt@goodmis.org> | 2026-09-02 11:08:21 -0400 |
| commit | 4617721c502b2ddaa4e324e86da4997edf738fa5 (patch) | |
| tree | 350aa83b47da0b574a814795392b20b0bed83cbc /Documentation/input/devices/rotary-encoder.rst | |
| download | linux-stable-4617721c502b2ddaa4e324e86da4997edf738fa5.tar.gz linux-stable-4617721c502b2ddaa4e324e86da4997edf738fa5.zip | |
ftrace: Synchronize the initialization of ftrace_opsgrafted
There's some internal state that ftrace_ops needs to have set, but since
it can be declared outside of the ftrace.c code, it calls
ftrace_ops_init() on the ops in every global function. The issue is that
if two tasks call it on the same ops at the same time it is possible to
have the initialization of one corrupt the initialization of the other
call.
Create a ops_mutex to use to synchronize every initialization of the
ftrace_ops. The mutex is taken within checking the ftrace_ops flag that
states it was initializied but the flag is checked again after the mutex
has been taken. Checking first outside the mutex allows it to shortcut
having to take the mutex. But then the check needs to be done again after
the mute is taken in case of races.
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260902095501.6b59af20@gandalf.local.home
Fixes: f04f24fb7e48d ("ftrace, kprobes: Fix a deadlock on ftrace_regex_lock")
Reported-by: sashiko-bot@kernel.org
Close: https://lore.kernel.org/all/20260829025528.49A831F000E9@smtp.kernel.org/
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Diffstat (limited to 'Documentation/input/devices/rotary-encoder.rst')
| -rw-r--r-- | Documentation/input/devices/rotary-encoder.rst | 135 |
1 files changed, 135 insertions, 0 deletions
diff --git a/Documentation/input/devices/rotary-encoder.rst b/Documentation/input/devices/rotary-encoder.rst new file mode 100644 index 000000000..5865748c1 --- /dev/null +++ b/Documentation/input/devices/rotary-encoder.rst @@ -0,0 +1,135 @@ +============================================================ +rotary-encoder - a generic driver for GPIO connected devices +============================================================ + +:Author: Daniel Mack <daniel@caiaq.de>, Feb 2009 + +Function +-------- + +Rotary encoders are devices which are connected to the CPU or other +peripherals with two wires. The outputs are phase-shifted by 90 degrees +and by triggering on falling and rising edges, the turn direction can +be determined. + +Some encoders have both outputs low in stable states, others also have +a stable state with both outputs high (half-period mode) and some have +a stable state in all steps (quarter-period mode). + +The phase diagram of these two outputs look like this:: + + _____ _____ _____ + | | | | | | + Channel A ____| |_____| |_____| |____ + + : : : : : : : : : : : : + __ _____ _____ _____ + | | | | | | | + Channel B |_____| |_____| |_____| |__ + + : : : : : : : : : : : : + Event a b c d a b c d a b c d + + |<-------->| + one step + + |<-->| + one step (half-period mode) + + |<>| + one step (quarter-period mode) + +For more information, please see + https://en.wikipedia.org/wiki/Rotary_encoder + + +Events / state machine +---------------------- + +In half-period mode, state a) and c) above are used to determine the +rotational direction based on the last stable state. Events are reported in +states b) and d) given that the new stable state is different from the last +(i.e. the rotation was not reversed half-way). + +Otherwise, the following apply: + +a) Rising edge on channel A, channel B in low state + This state is used to recognize a clockwise turn + +b) Rising edge on channel B, channel A in high state + When entering this state, the encoder is put into 'armed' state, + meaning that there it has seen half the way of a one-step transition. + +c) Falling edge on channel A, channel B in high state + This state is used to recognize a counter-clockwise turn + +d) Falling edge on channel B, channel A in low state + Parking position. If the encoder enters this state, a full transition + should have happened, unless it flipped back on half the way. The + 'armed' state tells us about that. + +Platform requirements +--------------------- + +As there is no hardware dependent call in this driver, the platform it is +used with must support gpiolib. Another requirement is that IRQs must be +able to fire on both edges. + + +Board integration +----------------- + +To use this driver in your system, register a platform_device with the +name 'rotary-encoder' and associate the IRQs and some specific platform +data with it. Because the driver uses generic device properties, this can +be done either via device tree, ACPI, or using static board files, like in +example below: + +:: + + /* board support file example */ + + #include <linux/input.h> + #include <linux/gpio/machine.h> + #include <linux/property.h> + + #define GPIO_ROTARY_A 1 + #define GPIO_ROTARY_B 2 + + static struct gpiod_lookup_table rotary_encoder_gpios = { + .dev_id = "rotary-encoder.0", + .table = { + GPIO_LOOKUP_IDX("gpio-0", + GPIO_ROTARY_A, NULL, 0, GPIO_ACTIVE_LOW), + GPIO_LOOKUP_IDX("gpio-0", + GPIO_ROTARY_B, NULL, 1, GPIO_ACTIVE_HIGH), + { }, + }, + }; + + static const struct property_entry rotary_encoder_properties[] = { + PROPERTY_ENTRY_U32("rotary-encoder,steps-per-period", 24), + PROPERTY_ENTRY_U32("linux,axis", ABS_X), + PROPERTY_ENTRY_U32("rotary-encoder,relative_axis", 0), + { }, + }; + + static const struct software_node rotary_encoder_node = { + .properties = rotary_encoder_properties, + }; + + static struct platform_device rotary_encoder_device = { + .name = "rotary-encoder", + .id = 0, + }; + + ... + + gpiod_add_lookup_table(&rotary_encoder_gpios); + device_add_software_node(&rotary_encoder_device.dev, &rotary_encoder_node); + platform_device_register(&rotary_encoder_device); + + ... + +Please consult device tree binding documentation to see all properties +supported by the driver. |
