blaster4385/linux-IllusionX/drivers/iio, branch v6.12.10

iio: imu: inv_icm42600: fix spi burst write not supported

2025-01-17T12:41:00Z

commit c0f866de4ce447bca3191b9cefac60c4b36a7922 upstream. Burst write with SPI is not working for all icm42600 chips. It was only used for setting user offsets with regmap_bulk_write. Add specific SPI regmap config for using only single write with SPI. Fixes: 9f9ff91b775b ("iio: imu: inv_icm42600: add SPI driver for inv_icm42600 driver") Cc: stable@vger.kernel.org Signed-off-by: Jean-Baptiste Maneyrol Link: https://patch.msgid.link/20241112-inv-icm42600-fix-spi-burst-write-not-supported-v2-1-97690dc03607@tdk.com Signed-off-by: Jonathan Cameron Signed-off-by: Greg Kroah-Hartman

iio: adc: ad7124: Disable all channels at probe time

2025-01-17T12:40:58Z

commit 4be339af334c283a1a1af3cb28e7e448a0aa8a7c upstream. When during a measurement two channels are enabled, two measurements are done that are reported sequencially in the DATA register. As the code triggered by reading one of the sysfs properties expects that only one channel is enabled it only reads the first data set which might or might not belong to the intended channel. To prevent this situation disable all channels during probe. This fixes a problem in practise because the reset default for channel 0 is enabled. So all measurements before the first measurement on channel 0 (which disables channel 0 at the end) might report wrong values. Fixes: 7b8d045e497a ("iio: adc: ad7124: allow more than 8 channels") Reviewed-by: Nuno Sa Signed-off-by: Uwe Kleine-König Link: https://patch.msgid.link/20241104101905.845737-2-u.kleine-koenig@baylibre.com Cc: Signed-off-by: Jonathan Cameron Signed-off-by: Greg Kroah-Hartman

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

2025-01-17T12:40:58Z

commit 36a44e05cd807a54e5ffad4b96d0d67f68ad8576 upstream. 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 Signed-off-by: Greg Kroah-Hartman

iio: inkern: call iio_device_put() only on mapped devices

2025-01-17T12:40:58Z

commit 64f43895b4457532a3cc524ab250b7a30739a1b1 upstream. In the error path of iio_channel_get_all(), iio_device_put() is called on all IIO devices, which can cause a refcount imbalance. Fix this error by calling iio_device_put() only on IIO devices whose refcounts were previously incremented by iio_device_get(). Fixes: 314be14bb893 ("iio: Rename _st_ functions to loose the bit that meant the staging version.") Signed-off-by: Joe Hattori Link: https://patch.msgid.link/20241204111342.1246706-1-joe@pf.is.s.u-tokyo.ac.jp Cc: Signed-off-by: Jonathan Cameron Signed-off-by: Greg Kroah-Hartman

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

2025-01-17T12:40:57Z

commit de6a73bad1743e9e81ea5a24c178c67429ff510b upstream. 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 Signed-off-by: Greg Kroah-Hartman

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

2025-01-17T12:40:57Z

commit 2a8e34096ec70d73ebb6d9920688ea312700cbd9 upstream. 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 Signed-off-by: Greg Kroah-Hartman

iio: imu: inv_icm42600: fix timestamps after suspend if sensor is on

2025-01-17T12:40:57Z

commit 65a60a590142c54a3f3be11ff162db2d5b0e1e06 upstream. Currently suspending while sensors are one will result in timestamping continuing without gap at resume. It can work with monotonic clock but not with other clocks. Fix that by resetting timestamping. Fixes: ec74ae9fd37c ("iio: imu: inv_icm42600: add accurate timestamping") Cc: stable@vger.kernel.org Signed-off-by: Jean-Baptiste Maneyrol Link: https://patch.msgid.link/20241113-inv_icm42600-fix-timestamps-after-suspend-v1-1-dfc77c394173@tdk.com Signed-off-by: Jonathan Cameron Signed-off-by: Greg Kroah-Hartman

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

2025-01-17T12:40:57Z

commit bcb394bb28e55312cace75362b8e489eb0e02a30 upstream. 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 Signed-off-by: Greg Kroah-Hartman

iio: gyro: fxas21002c: Fix missing data update in trigger handler

2025-01-17T12:40:57Z

commit fa13ac6cdf9b6c358e7d77c29fb60145c7a87965 upstream. The fxas21002c_trigger_handler() may fail to acquire sample data because the runtime PM enters the autosuspend state and sensor can not return sample data in standby mode.. Resume the sensor before reading the sample data into the buffer within the trigger handler. After the data is read, place the sensor back into the autosuspend state. Fixes: a0701b6263ae ("iio: gyro: add core driver for fxas21002c") Signed-off-by: Carlos Song Signed-off-by: Frank Li Link: https://patch.msgid.link/20241116152945.4006374-1-Frank.Li@nxp.com Cc: Signed-off-by: Jonathan Cameron Signed-off-by: Greg Kroah-Hartman

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

2025-01-17T12:40:57Z

commit 54d394905c92b9ecc65c1f9b2692c8e10716d8e1 upstream. 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 Signed-off-by: Greg Kroah-Hartman