In preparation to have dev->vals memory pre-allocated rearrange
code in input_alloc_device() so that it allows handling multiple
points of failure.
Reviewed-by: Jeff LaBundy <jeff@labundy.com>
Reviewed-by: Benjamin Tissoires <bentiss@kernel.org>
Link: https://lore.kernel.org/r/20240703213756.3375978-6-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Streamline event handling code by providing batch implementations for
filtering and event processing and using them in place of the main
event handler, as needed, instead of having complex branching logic
in the middle of the event processing code.
Reviewed-by: Jeff LaBundy <jeff@labundy.com>
Reviewed-by: Benjamin Tissoires <bentiss@kernel.org>
Link: https://lore.kernel.org/r/20240703213756.3375978-5-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

In preparation to consolidating filtering and event processing in the
input core change events() method to return number of events processed
by it.
Reviewed-by: Jeff LaBundy <jeff@labundy.com>
Reviewed-by: Benjamin Tissoires <bentiss@kernel.org>
Link: https://lore.kernel.org/r/20240703213756.3375978-4-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Input core expects input handlers to be either filters, or regular
handlers, but not both. Additionally, for regular handlers it does
not make sense to define both single event method and batch method.

Refuse registering handler if it defines more than one method.
Reviewed-by: Jeff LaBundy <jeff@labundy.com>
Reviewed-by: Benjamin Tissoires <bentiss@kernel.org>
Link: https://lore.kernel.org/r/20240703213756.3375978-3-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Input core favors ->events() (batch) method over ->event() method
if the former is defined, so there is no point in defining evdev_event()
as it is never called. Remove it.
Reviewed-by: Jeff LaBundy <jeff@labundy.com>
Reviewed-by: Benjamin Tissoires <bentiss@kernel.org>
Link: https://lore.kernel.org/r/20240703213756.3375978-2-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Imagis IST3038 is another variant of Imagis IST3038 IC, which has
a different register interface from IST3038C (possibly firmware defined).

Unlike IST3038C/IST3032C, IST3038 has different registers for commands,
which means IST3038 doesn't use protocol B.
Similar to IST3032C and maybe the other variants, IST3038 has touch keys
support, which provides KEY_APPSELECT and KEY_BACK.

Add support for IST3038 with touch keys.
Signed-off-by: Raymond Hackley <raymondhackley@protonmail.com>
Link: https://lore.kernel.org/r/20240613025631.5425-4-raymondhackley@protonmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Imagis IST3038 is a variant of Imagis touchscreen IC. Document it in
imagis,ist3038c bindings.
Signed-off-by: Raymond Hackley <raymondhackley@protonmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20240613025631.5425-3-raymondhackley@protonmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

protocol_b is a property, which tells Imagis panel to use a different
format for coordinates.

IST30XXC series is known for using protocol B, while the other series
aren't. Note this could be confusing, unlike the model name implies.

Adjust the usage of protocol_b to avoid confusion.
Signed-off-by: Raymond Hackley <raymondhackley@protonmail.com>
Link: https://lore.kernel.org/r/20240613025631.5425-2-raymondhackley@protonmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Convert TI-NSPIRE Keypad controller bindings to DT schema.
Signed-off-by: Andrew Davis <afd@ti.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20240612150711.26706-1-afd@ti.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

`mx25_tcq_regconfig` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Link: https://lore.kernel.org/r/20240705-input-const-regmap_config-v1-2-f712a4494883@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

`qt1050_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Link: https://lore.kernel.org/r/20240705-input-const-regmap_config-v1-1-f712a4494883@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

When dealing with byte data use u8 instead of unsigned char or int.
Stop layering error handling in cypress_ps2_sendbyte() and simply
pass on error code from ps2_sendbyte().

Additionally use u8 instead of unisgned char throughout the code.

Link: https://lore.kernel.org/r/20240628224728.2180126-5-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Do not override errors reported by lower layers with generic "-1",
but propagate them to the callers. Change the checks for errors to be
in the form of "if (error)" to maintain consistency.

Link: https://lore.kernel.org/r/20240628224728.2180126-4-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Report -ETIMEDOUT error code from cypress_ps2_read_cmd_status() when
device does not send enough data within the allotted time in response
to a command.

Link: https://lore.kernel.org/r/20240628224728.2180126-3-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Stop layering error handling in cypress_ps2_sendbyte() and simply
pass on error code from ps2_sendbyte() and use it in the callers.

This fixes mishandling of error condition in
cypress_ps2_read_cmd_status() which expects errors to be negative.
Reported-by: Igor Artemiev <Igor.A.Artemiev@mcst.ru>
Link: https://lore.kernel.org/r/20240628224728.2180126-2-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Casting an integer field containing desired rate to a pointer to bytes
works on little endian architectures where the driver is used, but not
a good practice. Use a temporary of proper type instead.

Link: https://lore.kernel.org/r/20240628224728.2180126-1-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Make sure interrupts are not left disabled when we fail to suspend the
touch controller.

Fixes: 6696777c ("Input: add driver for Elan I2C/SMbus touchpad")
Link: https://lore.kernel.org/r/ZmKiiL-1wzKrhqBj@google.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

This makes the code more compact and error handling more robust.

Link: https://lore.kernel.org/r/20240609234757.610273-3-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Start using __free() attributes to simplify the code and error handling.

Link: https://lore.kernel.org/r/20240609234757.610273-2-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Allocating a contiguous buffer of 64K may fail if memory is sufficiently
fragmented, and may cause OOM kill of an unrelated process. However we
do not need to have contiguous memory. We also do not need to zero
out the buffer since it will be overwritten with firmware data.

Switch to using kvmalloc() instead of kzalloc().

Link: https://lore.kernel.org/r/20240609234757.610273-1-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Start using __free() and guard() primitives to simplify the code
and error handling.

Link: https://lore.kernel.org/r/20240609235134.614592-3-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

sysfs_emit() is preferred over snprintf() for sysfs attribute handling.

Link: https://lore.kernel.org/r/20240609235134.614592-2-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

The code to toggle axis swapping and inversion is repetitive and can
be factored out.

Link: https://lore.kernel.org/r/20240609235134.614592-1-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Switch the driver to use __free(kfree) cleanup facility instead of
freeing memory by hand.

Link: https://lore.kernel.org/r/Zmk1SGwVt3rIbbMU@google.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Switch the driver to use guard notation when acquiring mutex to
have it released automatically.

Link: https://lore.kernel.org/r/Zmkyvkr9AFyywy1V@google.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Bring in ADS7846 hsync changes.

If the device is missing, we get the following error:

  qt1050 3-0041: ID -1340767592 not supported

Let's handle this situation and print more informative error
when reading of CHIP_ID fails:

  qt1050 3-0041: Failed to read chip ID: -6

Fixes: cbebf5ad ("Input: qt1050 - add Microchip AT42QT1050 support")
Signed-off-by: Andrei Lalaev <andrei.lalaev@anton-paar.com>
Reviewed-by: Marco Felsch <m.felsch@pengutronix.de>
Link: https://lore.kernel.org/r/20240617183018.916234-1-andrey.lalaev@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

As ft5426 seems to be compatible with this driver, add it.
Debug output during identification: Model "generic ft5x06 (79)", Rev. "
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Reviewed-by: Oliver Graute <oliver.graute@kococonnector.com>
Link: https://lore.kernel.org/r/20240501204758.758537-3-andreas@kemnade.infoSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Add compatible for ft5426.
Searches for documentation reveal neither edt nor evervision
as some related company, only FocalTech.
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Oliver Graute <oliver.graute@kococonnector.com>
Link: https://lore.kernel.org/r/20240501204758.758537-2-andreas@kemnade.infoSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

The chip is similar, but has status bits at different positions,
so use the correct bits.
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Link: https://lore.kernel.org/r/20240621224022.1620897-4-andreas@kemnade.infoSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Add a compatible for the EKTF2232, which is similar to other chips in this
document.
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20240621224022.1620897-3-andreas@kemnade.infoSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Convert EKTF2127 infrared touchscreen controller binding to DT schema
and add ektf2232 compatible.
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20240621224022.1620897-2-andreas@kemnade.infoSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Add handling of HSYNC signal emitted by the TI ADS7846 if it is hooked
up to a GPIO.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20240430-gpio-leds-miscarm-v1-3-9c94d7711f6c@linaro.org
[dtor: dropped Spitz changes, kept platform data wait_for_sync option]
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

The TI ADS7846 emits a horizontal sync signal that is usually
connected to a GPIO for polling. Add a binding for this.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20240430-gpio-leds-miscarm-v1-2-9c94d7711f6c@linaro.orgSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

There is no need to allocate axes information separately from the main
joystick structure so let's fold the allocation and also drop members
(such as range, flat and fuzz) that are only used during initialization
of the device.
Acked-by: Artur Rojek <contact@artur-rojek.eu>
Link: https://lore.kernel.org/r/ZmkrgTlxNwm_oHxv@google.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Start using __free() and guard() primitives to simplify the code
and error handling.

Link: https://lore.kernel.org/r/20240610041813.722445-2-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Instead of manually creating driver-specific device attributes
set struct usb_driver->dev_groups pointer to have the driver core
do it.

Link: https://lore.kernel.org/r/20240610041813.722445-1-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

This case of the common error message upon failure of
input_allocate_device() repeats the word "input".

Drop one "input" from the error message.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Kees Cook <kees@kernel.org>
Link: https://lore.kernel.org/r/20240619-ims-pcu-counted_by-v1-2-3ee0ead2e57d@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Use the __counted_by compiler attribute for the data[] flexible array
member to improve the results of array bound sanitizers.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Kees Cook <kees@kernel.org>
Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20240619-ims-pcu-counted_by-v1-1-3ee0ead2e57d@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Switch to using guard(mutex)() notation to acquire and automatically
release mutexes.
Reviewed-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240610164301.1048482-4-dmitry.torokhov@gmail.comSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>