linux.git/drivers/iio/adc, branch for-next

iio: adc: ti-ads1119: fix sample size in scan struct for triggered buffer

2024-12-08T16:59:13+00:00

This device returns signed, 16-bit samples as stated in its datasheet
(see 8.5.2 Data Format). That is in line with the scan_type definition
for the IIO_VOLTAGE channel, but 'unsigned int' is being used to read
and push the data to userspace.

Given that the size of that type depends on the architecture (at least
2 bytes to store values up to 65535, but its actual size is often 4
bytes), use the 's16' type to provide the same structure in all cases.

Fixes: a9306887eba4 ("iio: adc: ti-ads1119: Add driver")
Signed-off-by: Javier Carrasco 
Reviewed-by: Francesco Dolcini 
Link: https://patch.msgid.link/20241202-ti-ads1119_s16_chan-v1-1-fafe3136dc90@gmail.com
Cc: 
Signed-off-by: Jonathan Cameron

iio: adc: ad9467: Fix the "don't allow reading vref if not available" case

2024-12-07T17:57:19+00:00

The commit in Fixes adds a special case when only one possible scale is
available.
If several scales are available, it sets the .read_avail field of the
struct iio_info to ad9467_read_avail().

However, this field already holds this function pointer, so the code is a
no-op.

Use another struct iio_info instead to actually reflect the intent
described in the commit message. This way, the structure to use is selected
at runtime and they can be kept as const.

This is safer because modifying static structs that are shared between all
instances like this, based on the properties of a single instance, is
asking for trouble down the road.

Fixes: b92f94f74826 ("iio: adc: ad9467: don't allow reading vref if not available")
Signed-off-by: Christophe JAILLET 
Link: https://patch.msgid.link/cc65da19e0578823d29e11996f86042e84d5715c.1733503146.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Jonathan Cameron

iio: adc: at91: call input_free_device() on allocated iio_dev

2024-12-07T17:28:39+00:00

Current implementation of at91_ts_register() calls input_free_deivce()
on st->ts_input, however, the err label can be reached before the
allocated iio_dev is stored to st->ts_input. Thus call
input_free_device() on input instead of st->ts_input.

Fixes: 84882b060301 ("iio: adc: at91_adc: Add support for touchscreens without TSMR")
Signed-off-by: Joe Hattori 
Link: https://patch.msgid.link/20241207043045.1255409-1-joe@pf.is.s.u-tokyo.ac.jp
Cc: 
Signed-off-by: Jonathan Cameron

iio: adc: ad7173: fix using shared static info struct

2024-12-07T17:23:42+00:00

Fix a possible race condition during driver probe in the ad7173 driver
due to using a shared static info struct. If more that one instance of
the driver is probed at the same time, some of the info could be
overwritten by the other instance, leading to incorrect operation.

To fix this, make the static info struct const so that it is read-only
and make a copy of the info struct for each instance of the driver that
can be modified.

Reported-by: Uwe Kleine-König 
Fixes: 76a1e6a42802 ("iio: adc: ad7173: add AD7173 driver")
Signed-off-by: David Lechner 
Tested-by: Guillaume Ranquet 
Link: https://patch.msgid.link/20241127-iio-adc-ad7313-fix-non-const-info-struct-v2-1-b6d7022b7466@baylibre.com
Cc: 
Signed-off-by: Jonathan Cameron

iio: adc: ti-ads124s08: Use gpiod_set_value_cansleep()

2024-12-07T17:23:42+00:00

Using gpiod_set_value() to control the reset GPIO causes some verbose
warnings during boot when the reset GPIO is controlled by an I2C IO
expander.

As the caller can sleep, use the gpiod_set_value_cansleep() variant to
fix the issue.

Tested on a custom i.MX93 board with a ADS124S08 ADC.

Cc: stable@kernel.org
Fixes: e717f8c6dfec ("iio: adc: Add the TI ads124s08 ADC code")
Signed-off-by: Fabio Estevam 
Link: https://patch.msgid.link/20241122164308.390340-1-festevam@gmail.com
Signed-off-by: Jonathan Cameron

iio: adc: ti-ads1119: fix information leak in triggered buffer

2024-12-07T17:23:42+00:00

The 'scan' local struct is used to push data to user space from a
triggered buffer, but it has a hole between the sample (unsigned int)
and the timestamp. This hole is never initialized.

Initialize the struct to zero before using it to avoid pushing
uninitialized information to userspace.

Cc: stable@vger.kernel.org
Fixes: a9306887eba4 ("iio: adc: ti-ads1119: Add driver")
Signed-off-by: Javier Carrasco 
Reviewed-by: Francesco Dolcini 
Link: https://patch.msgid.link/20241125-iio_memset_scan_holes-v1-2-0cb6e98d895c@gmail.com
Signed-off-by: Jonathan Cameron

iio: adc: rockchip_saradc: fix information leak in triggered buffer

2024-12-07T17:23:42+00:00

The 'data' local struct is used to push data to user space from a
triggered buffer, but it does not set values for inactive channels, as
it only uses iio_for_each_active_channel() to assign new values.

Initialize the struct to zero before using it to avoid pushing
uninitialized information to userspace.

Cc: stable@vger.kernel.org
Fixes: 4e130dc7b413 ("iio: adc: rockchip_saradc: Add support iio buffers")
Signed-off-by: Javier Carrasco 
Link: https://patch.msgid.link/20241125-iio_memset_scan_holes-v1-4-0cb6e98d895c@gmail.com
Signed-off-by: Jonathan Cameron

iio: adc: ti-ads8688: fix information leak in triggered buffer

2024-12-07T17:23:41+00:00

The 'buffer' local array is used to push data to user space from a
triggered buffer, but it does not set values for inactive channels, as
it only uses iio_for_each_active_channel() to assign new values.

Initialize the array to zero before using it to avoid pushing
uninitialized information to userspace.

Cc: stable@vger.kernel.org
Fixes: 61fa5dfa5f52 ("iio: adc: ti-ads8688: Fix alignment of buffer in iio_push_to_buffers_with_timestamp()")
Signed-off-by: Javier Carrasco 
Link: https://patch.msgid.link/20241125-iio_memset_scan_holes-v1-8-0cb6e98d895c@gmail.com
Signed-off-by: Jonathan Cameron

iio: adc: ti-ads1298: Add NULL check in ads1298_init

2024-12-07T17:23:41+00:00

devm_kasprintf() can return a NULL pointer on failure. A check on the
return value of such a call in ads1298_init() is missing. Add it.

Fixes: 00ef7708fa60 ("iio: adc: ti-ads1298: Add driver")
Signed-off-by: Charles Han 
Link: https://patch.msgid.link/20241118090208.14586-1-hanchunchao@inspur.com
Cc: 
Signed-off-by: Jonathan Cameron

iio: adc: stm32-dfsdm: handle label as an optional property

2024-12-07T17:23:41+00:00

The label property is defined as optional in the DFSDM binding.
Parse the label property only when it is defined in the device tree.

Fixes: 3208fa0cd919 ("iio: adc: stm32-dfsdm: adopt generic channels bindings")
Signed-off-by: Olivier Moysan 
Link: https://patch.msgid.link/20241114102459.2497178-1-olivier.moysan@foss.st.com
Signed-off-by: Jonathan Cameron