]> git.ipfire.org Git - thirdparty/kernel/linux.git/log
thirdparty/kernel/linux.git
2 months agoiio: proximity: add driver for ST VL53L1X ToF sensor
Siratul Islam [Wed, 25 Mar 2026 20:19:42 +0000 (02:19 +0600)] 
iio: proximity: add driver for ST VL53L1X ToF sensor

Add support for the STMicroelectronics VL53L1X Time-of-Flight
ranging sensor with I2C interface.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Siratul Islam <email@sirat.me>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agodt-bindings: iio: proximity: add ST VL53L1X ToF sensor
Siratul Islam [Wed, 25 Mar 2026 20:19:41 +0000 (02:19 +0600)] 
dt-bindings: iio: proximity: add ST VL53L1X ToF sensor

Add device tree binding documentation for the STMicroelectronics
VL53L1X Time-of-Flight ranging sensor connected via I2C.

vdd-supply is not made globally required to maintain backwards
compatibility with existing st,vl53l0x devicetrees that do not specify it.

Signed-off-by: Siratul Islam <email@sirat.me>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: max11410: make vref register name arrays static const
Giorgi Tchankvetadze [Mon, 23 Mar 2026 06:46:41 +0000 (10:46 +0400)] 
iio: adc: max11410: make vref register name arrays static const

The vrefp_regs and vrefn_regs arrays are constant lookup tables and
are not modified.

Make them static const so they are not reinitialized on each probe
call and are placed in read-only memory. Mark the pointer array as
const as well to prevent unintended modification.

Signed-off-by: Giorgi Tchankvetadze <giorgitchankvetadze1997@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: accel: bmc150-accel-core: use sysfs_emit() in show functions
Gabriel Rondon [Mon, 23 Mar 2026 21:56:19 +0000 (21:56 +0000)] 
iio: accel: bmc150-accel-core: use sysfs_emit() in show functions

Replace sprintf() with sysfs_emit() in sysfs attribute show
callbacks. sysfs_emit() is the preferred API as it is aware of the
sysfs buffer page size limit.

Signed-off-by: Gabriel Rondon <grondon@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: frequency: adf4350: replace TODO with NOTE in adf4350_set_freq()
Neel Bullywon [Mon, 23 Mar 2026 23:33:16 +0000 (19:33 -0400)] 
iio: frequency: adf4350: replace TODO with NOTE in adf4350_set_freq()

Replace the TODO comment in adf4350_set_freq() with a NOTE explaining
that a constant-time approach using fls_long() was attempted but deemed
more complex without meaningful benefit for initialization code.

Signed-off-by: Neel Bullywon <neelb2403@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: ltc2309: add support for ltc2305
Kyle Hsieh [Wed, 25 Mar 2026 02:24:22 +0000 (10:24 +0800)] 
iio: adc: ltc2309: add support for ltc2305

Add support for the LTC2305 ADC to the LTC2309 driver. The LTC2305 is
a 2-channel, 12-bit SAR ADC that is register-compatible with the
LTC2309 but has a different channel selection mapping and count.

To support multiple chips in this family, introduce ltc2309_chip_info
struct to store chip-specific channel specifications and names.
The probe function now uses i2c_get_match_data() to retrieve the
correct configuration for the detected device.

Specific channel addresses for LTC2305 (CH0, CH1, and differential
pairs) are added based on the datasheet.

Signed-off-by: Kyle Hsieh <kylehsieh1995@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: ltc2309: explicitly assign hex values to channel enums
Kyle Hsieh [Wed, 25 Mar 2026 02:24:21 +0000 (10:24 +0800)] 
iio: adc: ltc2309: explicitly assign hex values to channel enums

The current ltc2309_channels enum relies on implicit sequential
assignment. While this works for the 8-channel LTC2309, it is
not intuitive and makes it difficult to support other chips in
the same family that might have different bit mappings.

Explicitly assign hex values to the enum members based on the
channel selection bits defined in the datasheet. This improves
code readability and provides a consistent pattern for future
chip support.

Signed-off-by: Kyle Hsieh <kylehsieh1995@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agodt-bindings: adc: ltc2497: add support for ltc2305
Kyle Hsieh [Wed, 25 Mar 2026 02:24:20 +0000 (10:24 +0800)] 
dt-bindings: adc: ltc2497: add support for ltc2305

Add documentation for the 2-channel LTC2305 ADC in the
existing ltc2497 binding.
This enables automatic device tree matching for LTC2305
while using the LTC2309 driver (drivers/iio/adc/ltc2309.c),
since both ADCs share the same I2C interface and 12-bit SAR architecture.
The main difference is the number of channels (LTC2305: 2, LTC2309: 8).

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Kyle Hsieh <kylehsieh1995@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: accel: adxl380: fix typo in PART_ID register macro
Nikhil Gautam [Wed, 25 Mar 2026 11:42:51 +0000 (17:12 +0530)] 
iio: accel: adxl380: fix typo in PART_ID register macro

Fix a typo in the ADXL380_PART_ID_REG macro name where it was
incorrectly defined as ADLX380_PART_ID_REG.

Also update its usage in adxl380_setup().

Signed-off-by: Nikhil Gautam <nikhilgtr@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: add Rfs-based scale and per-variant limits
Oleksij Rempel [Tue, 10 Feb 2026 13:51:10 +0000 (14:51 +0100)] 
iio: dac: ds4424: add Rfs-based scale and per-variant limits

Parse optional maxim,rfs-ohms values to derive the per-channel output
current scale (mA per step) for the IIO current ABI.

Behavior changes:
- If maxim,rfs-ohms is present, IIO_CHAN_INFO_SCALE becomes available
  and reports mA/step derived from Rfs.
- If maxim,rfs-ohms is missing, SCALE is not exposed to keep older DTs
  working without requiring updates.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agodt-bindings: iio: dac: maxim,ds4424: add maxim,rfs-ohms property
Oleksij Rempel [Tue, 10 Feb 2026 13:51:09 +0000 (14:51 +0100)] 
dt-bindings: iio: dac: maxim,ds4424: add maxim,rfs-ohms property

The Maxim DS4422/DS4424 and DS4402/DS4404 current DACs determine their
full-scale output current via external resistors (Rfs) connected to the
FSx pins. Without knowing these values, the full-scale range of the
hardware is undefined.

Add the 'maxim,rfs-ohms' property to describe these physical components.
This property is required to provide a complete description of the
hardware configuration.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: convert to regmap
Oleksij Rempel [Tue, 10 Feb 2026 13:51:08 +0000 (14:51 +0100)] 
iio: dac: ds4424: convert to regmap

Refactor the driver to use the regmap API.

Replace the driver-specific mutex and manual shadow buffers with the
standard regmap infrastructure for locking and caching.

This ensures the cache is populated from hardware at probe, preventing
state desynchronization (e.g. across suspend/resume).

Define access tables to validate the different register maps of DS44x2
and DS44x4.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Sander Vanheule <sander@svanheule.net>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: support per-variant output range limits
Oleksij Rempel [Tue, 10 Feb 2026 13:51:07 +0000 (14:51 +0100)] 
iio: dac: ds4424: support per-variant output range limits

The DS4402/DS4404 variants operate with a 5-bit resolution (31 steps),
whereas the DS4422/DS4424 support 7-bit (127 steps).

Previously, the driver enforced a hardcoded 7-bit mask (DS4424_DAC_MASK)
for all variants. This allowed users to write values exceeding the 5-bit
range to DS4402/DS4404 devices, resulting in silent truncation or
undefined behavior.

Add a `result_mask` field to the chip_info structure to define the valid
data range for each variant. Use this mask to:
1. Correctly mask register values in read_raw().
2. Return -EINVAL in write_raw() if the input value exceeds the
   variant's capabilities.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: add DS4402/DS4404 device IDs
David Jander [Tue, 10 Feb 2026 13:51:06 +0000 (14:51 +0100)] 
iio: dac: ds4424: add DS4402/DS4404 device IDs

Add I2C/OF IDs for DS4402 and DS4404 and set the correct channel count.
Follow-up changes add per-variant scaling based on external Rfs.

Co-developed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David Jander <david@protonic.nl>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agodt-bindings: iio: dac: maxim,ds4424: add ds4402/ds4404
Oleksij Rempel [Tue, 10 Feb 2026 13:51:05 +0000 (14:51 +0100)] 
dt-bindings: iio: dac: maxim,ds4424: add ds4402/ds4404

Add compatible strings for Maxim DS4402 and DS4404 current DACs.
These devices are 5-bit variants of the DS4422/DS4424 family.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: use fsleep() instead of usleep_range()
Oleksij Rempel [Tue, 10 Feb 2026 13:51:04 +0000 (14:51 +0100)] 
iio: dac: ds4424: use fsleep() instead of usleep_range()

The DS4422/DS4424 and DS4402/DS4404 datasheets do not specify a minimum
delay between power-up (POR) and the availability of the I2C interface.

The driver previously used `usleep_range(1000, 1200)` to enforce a ~1ms
delay. Replace this with `fsleep(1000)` to allow the kernel to select
the most efficient sleep mechanism while retaining the existing
conservative delay to ensure device readiness.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: use device match data for chip info
Oleksij Rempel [Tue, 10 Feb 2026 13:51:03 +0000 (14:51 +0100)] 
iio: dac: ds4424: use device match data for chip info

Refactor the driver to use device match data instead of checking ID enums
in a switch statement.

Define a `ds4424_chip_info` structure to hold variant-specific attributes
(currently just the channel count) and attach it directly to the I2C and
OF device ID tables.

This simplifies the probe function and makes it easier to add support for
new variants like DS4402/DS4404.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: rename iio_info struct to avoid ambiguity
Oleksij Rempel [Tue, 10 Feb 2026 13:51:02 +0000 (14:51 +0100)] 
iio: dac: ds4424: rename iio_info struct to avoid ambiguity

Rename the static `ds4424_info` structure to `ds4424_iio_info`.

The previous name was generic and could be confused with chip-specific
data structures (like the upcoming `ds4424_chip_info`). The new name
explicitly indicates that this structure holds the IIO framework
callbacks.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: sort headers alphabetically
Oleksij Rempel [Tue, 10 Feb 2026 13:51:01 +0000 (14:51 +0100)] 
iio: dac: ds4424: sort headers alphabetically

Sort the header inclusions alphabetically. This improves readability and
simplifies adding new includes in the future.

Group subsystem-specific headers (linux/iio/*) separately at the end
to clarify subsystem context.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: ratelimit read errors and use device context
Oleksij Rempel [Tue, 10 Feb 2026 13:51:00 +0000 (14:51 +0100)] 
iio: dac: ds4424: ratelimit read errors and use device context

Replace pr_err() with dev_err_ratelimited() in the RAW read path to avoid
log spam on repeated I2C failures and to include the device context.

Use %pe to print errno names for faster debugging. Use the parent
device context to identify the physical hardware causing the error.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ds4424: refactor raw access to use bitwise operations
Oleksij Rempel [Tue, 10 Feb 2026 13:50:59 +0000 (14:50 +0100)] 
iio: dac: ds4424: refactor raw access to use bitwise operations

Refactor the raw access logic to use standard GENMASK() and BIT()
macros. Use abs() for magnitude calculation to simplify the logic and
make the data flow clearer.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoMerge tag 'v7.0-rc4' into togreg
Jonathan Cameron [Sun, 22 Mar 2026 12:20:42 +0000 (12:20 +0000)] 
Merge tag 'v7.0-rc4' into togreg

Linux 7.0-rc4

Required for the ds4422 series which is build upon;
5187e03b817c ("iio: dac: ds4424: reject -128 RAW value")

2 months agoiio: st_sensors: correct kernel-doc issues
Randy Dunlap [Thu, 12 Mar 2026 05:13:45 +0000 (22:13 -0700)] 
iio: st_sensors: correct kernel-doc issues

Use the proper kernel-doc format and struct member names to avoid
kernel-doc warnings:

Warning: include/linux/iio/common/st_sensors.h:184 struct member 'int1'
 not described in 'st_sensor_data_ready_irq'
Warning: ../include/linux/iio/common/st_sensors.h:184 struct member 'int2'
 not described in 'st_sensor_data_ready_irq'
Warning: ../include/linux/iio/common/st_sensors.h:184 struct member
 'stat_drdy' not described in 'st_sensor_data_ready_irq'
Warning: ../include/linux/iio/common/st_sensors.h:184 struct member 'ig1'
 not described in 'st_sensor_data_ready_irq'
Warning: ../include/linux/iio/common/st_sensors.h:219 struct member
 'num_ch' not described in 'st_sensor_settings'
Warning: ../include/linux/iio/common/st_sensors.h:263 struct member
 'num_data_channels' not described in 'st_sensor_data'

Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: frequency: ad9523: use dev_err_probe
Antoniu Miclaus [Fri, 13 Mar 2026 11:57:44 +0000 (13:57 +0200)] 
iio: frequency: ad9523: use dev_err_probe

Use dev_err_probe() instead of dev_err() in the probe path to ensure
proper handling of deferred probing and to simplify error handling.

Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: frequency: ad9523: add dev variable
Antoniu Miclaus [Fri, 13 Mar 2026 11:57:43 +0000 (13:57 +0200)] 
iio: frequency: ad9523: add dev variable

Introduce a local struct device variable in ad9523_probe() to simplify
subsequent conversions and improve code readability.

Split pdata declaration and assignment since the result is validated
immediately after.

No functional change.

Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: frequency: admv4420: use dev_err_probe
Antoniu Miclaus [Fri, 13 Mar 2026 11:57:42 +0000 (13:57 +0200)] 
iio: frequency: admv4420: use dev_err_probe

Use dev_err_probe() instead of dev_err() in the probe path to ensure
proper handling of deferred probing and to simplify error handling.

Also fix the format specifier for vco_freq_hz from %lld to %llu since
it is u64 (unsigned), and add missing newline to the error message.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: frequency: admv4420: add dev variable
Antoniu Miclaus [Fri, 13 Mar 2026 11:57:41 +0000 (13:57 +0200)] 
iio: frequency: admv4420: add dev variable

Introduce a local struct device variable in admv4420_probe() to
simplify subsequent conversions and improve code readability.

No functional change.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: ad_sigma_delta: Format block comments
Giorgi Tchankvetadze [Sat, 14 Mar 2026 11:12:53 +0000 (15:12 +0400)] 
iio: adc: ad_sigma_delta: Format block comments

Format the multi-line comment in ad_sd_set_comm() according to
the kernel multi-line comment style.

Suggested-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Giorgi Tchankvetadze <giorgitchankvetadze1997@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: ti-ads7950: use iio_push_to_buffers_with_ts_unaligned()
David Lechner [Sat, 14 Mar 2026 21:12:24 +0000 (16:12 -0500)] 
iio: adc: ti-ads7950: use iio_push_to_buffers_with_ts_unaligned()

Use iio_push_to_buffers_with_ts_unaligned() to avoid unaligned access
when writing the timestamp in the rx_buf.

The previous implementation would have been fine on architectures that
support 4-byte alignment of 64-bit integers but could cause issues on
architectures that require 8-byte alignment.

Fixes: 902c4b2446d4 ("iio: adc: New driver for TI ADS7950 chips")
Signed-off-by: David Lechner <dlechner@baylibre.com>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: st_sensors: drop temporary kmalloc buffer and reuse buffer_data
Sanjay Chitroda [Sun, 15 Mar 2026 12:16:25 +0000 (17:46 +0530)] 
iio: st_sensors: drop temporary kmalloc buffer and reuse buffer_data

Replace the per-call kmalloc() scratch buffer with the existing
buffer_data[] field present in struct st_sensor_data. The existing buffer
is DMA-aligned and sufficiently sized for all channel widths, so using it
avoids unnecessary dynamic memory allocation on each read.

This simplifies the code, removes redundant allocation and cleanup.
No functional change intended.

Signed-off-by: Sanjay Chitroda <sanjayembeddedse@gmail.com>
Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: ina2xx: add INA236 support
Chuang Zhu [Sun, 15 Mar 2026 18:23:04 +0000 (02:23 +0800)] 
iio: adc: ina2xx: add INA236 support

The calibration divisor is not directly specified in the datasheet, but can be calculated:

I = Current_LSB * Current
Current = ShuntVoltage * CAL / calibration_divisor
CAL = 0.00512 / (Current_LSB * Rshunt)
ShuntVoltage = Vshunt / ShuntVoltage_LSB

=> I = (0.00512 / (calibration_divisor*ShuntVoltage_LSB)) * (Vshunt / Rshunt)

Ohm's law, I = Vshunt / Rshunt
=> 0.00512 / (calibration_divisor*ShuntVoltage_LSB) = 1

ShuntVoltage_LSB = 2.5 uV = 0.0000025 V
=> calibration_divisor = 2048

Signed-off-by: Chuang Zhu <git@chuang.cz>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: as73211: remove duplicate zero init of scan.chan[3]
David Lechner [Sun, 15 Mar 2026 19:21:51 +0000 (14:21 -0500)] 
iio: light: as73211: remove duplicate zero init of scan.chan[3]

Remove setting scan.chan[3] to zero. Since commit 433b99e92294 ("iio:
light: as73211: Ensure buffer holes are zeroed"), the entire scan struct
is zeroed before being filled with data, so this is redundant.

Signed-off-by: David Lechner <dlechner@baylibre.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: amplifiers: fix typo from Curren to Current
Shi Hao [Mon, 16 Mar 2026 09:00:22 +0000 (14:30 +0530)] 
iio: amplifiers: fix typo from Curren to Current

Fix incorrect spelling from Curren to Current.

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: cdc: fix spelling mistakes in comments
Shi Hao [Mon, 16 Mar 2026 09:00:21 +0000 (14:30 +0530)] 
iio: cdc: fix spelling mistakes in comments

Fix spelling mistakes in comments.

- becaue -> because
- reenable -> re-enable
- irq's -> IRQs

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: chemical: rephrase comment and fix a typo
Shi Hao [Mon, 16 Mar 2026 09:00:20 +0000 (14:30 +0530)] 
iio: chemical: rephrase comment and fix a typo

Rephrase the comment and fix a spelling mistake.

- insuffient -> insufficient

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: common: fix spelling mistakes in comments
Shi Hao [Mon, 16 Mar 2026 09:00:19 +0000 (14:30 +0530)] 
iio: common: fix spelling mistakes in comments

Fix spelling mistakes in comments.

- exepects -> expects
- fuction -> function
- theoritical -> theoretical
- appopriate -> appropriate
- iio -> IIO

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: test: fix typo from neeeds to needs in comment
Shi Hao [Mon, 16 Mar 2026 09:00:18 +0000 (14:30 +0530)] 
iio: test: fix typo from neeeds to needs in comment

Fix incorrect spelling from neeeds to needs.

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Reviewed-by: Matti Vaittinen <mazziesaccount@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: temp: fix spelling mistakes in comments
Shi Hao [Mon, 16 Mar 2026 09:00:17 +0000 (14:30 +0530)] 
iio: temp: fix spelling mistakes in comments

Fix spelling mistakes in comments.

- catched -> caught
- chanel -> channel

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: resolver: fix typo from degredation to degradation
Shi Hao [Mon, 16 Mar 2026 09:00:16 +0000 (14:30 +0530)] 
iio: resolver: fix typo from degredation to degradation

Fix incorrect spelling from degredation to degradation and fixed
up some missing spaces prior to */

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: proximity: fix typo from currenly to currently
Shi Hao [Mon, 16 Mar 2026 09:00:15 +0000 (14:30 +0530)] 
iio: proximity: fix typo from currenly to currently

Fix incorrect spelling from currenly to currently.

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: pressure: fix spelling mistakes in comments
Shi Hao [Mon, 16 Mar 2026 09:00:14 +0000 (14:30 +0530)] 
iio: pressure: fix spelling mistakes in comments

Fix several spelling mistakes in comments.

- opertion -> operations
- transfered -> transferred
- usng -> using
- externaly -> externally

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Reviewed-by: Matti Vaittinen <mazziesaccount@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: magnetometer: fix various spelling mistakes
Shi Hao [Mon, 16 Mar 2026 09:00:13 +0000 (14:30 +0530)] 
iio: magnetometer: fix various spelling mistakes

Fix spelling mistakes in comments.

- follwing -> following
- atleast -> at least
- occured -> occurred
- measurment -> measurement
- rougly -> roughly

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: imu: fix typo from adjustement to adjustment
Shi Hao [Mon, 16 Mar 2026 09:00:12 +0000 (14:30 +0530)] 
iio: imu: fix typo from adjustement to adjustment

Fix incorrect spelling in a comment.

- adjustement -> adjustment

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: add an article and use digitize instead of digitalize
Shi Hao [Mon, 16 Mar 2026 09:00:11 +0000 (14:30 +0530)] 
iio: adc: add an article and use digitize instead of digitalize

Use digitize instead of digitalize, which is the correct technical term,
and add an article for clarity.

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: fix several incorrect spellings
Shi Hao [Mon, 16 Mar 2026 09:00:10 +0000 (14:30 +0530)] 
iio: light: fix several incorrect spellings

Fix spelling mistakes reported by codespell.

- sesnor -> sensor
- substraction -> subtraction
- simulataneous -> simultaneous
- proccessed -> processed
- coefficents -> coefficients

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: accel: fix typo celcius to Celsius
Shi Hao [Mon, 16 Mar 2026 09:00:09 +0000 (14:30 +0530)] 
iio: accel: fix typo celcius to Celsius

Fix incorrect spelling in comments
- celcius -> Celsius

Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: aspeed: Reserve battery sensing channel for on-demand use
Billy Tsai [Fri, 20 Mar 2026 05:46:38 +0000 (13:46 +0800)] 
iio: adc: aspeed: Reserve battery sensing channel for on-demand use

For controllers with battery sensing capability (AST2600/AST2700), the
last channel uses a different circuit design optimized for battery
voltage measurement. This channel should not be enabled by default
along with other channels to avoid potential interference and power
efficiency issues.
This ensures optimal power efficiency for normal ADC operations while
maintaining full functionality when battery sensing is needed.

Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: aspeed: Replace mdelay() with fsleep() for ADC stabilization delay
Billy Tsai [Fri, 20 Mar 2026 05:46:37 +0000 (13:46 +0800)] 
iio: adc: aspeed: Replace mdelay() with fsleep() for ADC stabilization delay

The ADC stabilization delays in compensation mode and battery sensing
mode do not require atomic context. Using mdelay() here results in
unnecessary busy waiting.

Replace mdelay(1) with fsleep(1000) to allow the scheduler to run other
tasks while waiting for the ADC to stabilize.

Also fix a minor typo in the comment ("adc" -> "ADC").

Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: Enable multiple consecutive channels based on model data
Billy Tsai [Fri, 20 Mar 2026 05:46:36 +0000 (13:46 +0800)] 
iio: adc: Enable multiple consecutive channels based on model data

Add helpers to generate channel masks and enable multiple ADC channels
according to the device model's channel count.

Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: Add battery channel definition for ADC
Billy Tsai [Fri, 20 Mar 2026 05:46:35 +0000 (13:46 +0800)] 
iio: adc: Add battery channel definition for ADC

Defines a constant for the battery sensing channel, typically the last
channel of the ADC. Clarifies channel usage and improves code
readability.

Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ltc2632: add support for LTC2654 DAC family
David Marinovic [Fri, 20 Mar 2026 15:09:48 +0000 (16:09 +0100)] 
iio: dac: ltc2632: add support for LTC2654 DAC family

Add support for the Linear Technology LTC2654 quad DAC family.

The LTC2654 is a 4-channel, 16-/12-bit DAC with SPI interface,
sharing the same 24-bit SPI protocol as the existing LTC2632/
LTC2634/LTC2636 devices supported by this driver.

The 12-bit variants of LTC2654 reuse existing LTC2634 chip_info
structs as they are register-compatible.

Add support for the following variants:
- LTC2654L-16: 16-bit, 2.5V internal reference
- LTC2654L-12: 12-bit, 2.5V internal reference
- LTC2654H-16: 16-bit, 4.096V internal reference
- LTC2654H-12: 12-bit, 4.096V internal reference

Signed-off-by: David Marinovic <david.marinovic@pupin.rs>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agodt-bindings: iio: dac: ltc2632: add LTC2654 compatible strings
David Marinovic [Fri, 20 Mar 2026 15:09:47 +0000 (16:09 +0100)] 
dt-bindings: iio: dac: ltc2632: add LTC2654 compatible strings

Add compatible strings for the LTC2654 quad-channel DAC family.

The LTC2654 devices are 4-channel, 16-/12-bit DACs with an internal
reference and SPI interface. They use the same 24-bit SPI command
format as the LTC2632/2634/2636 family.

The 16-bit variants (LTC2654-L16 and LTC2654-H16) require new
compatible strings, as no existing compatibles support 16-bit
resolution.

The 12-bit variants (LTC2654-L12 and LTC2654-H12) are register-
compatible with LTC2634-L12 and LTC2634-H12 respectively, and can
use them as fallback compatibles.

Signed-off-by: David Marinovic <david.marinovic@pupin.rs>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: dac: ltc2632: drop enum and use individual chip_info objects
David Marinovic [Fri, 20 Mar 2026 15:09:46 +0000 (16:09 +0100)] 
iio: dac: ltc2632: drop enum and use individual chip_info objects

Remove the ltc2632_chip_info_tbl[] array and related
ltc2632_supported_device_ids enum used for looking up chip-specific
information. Instead, use separate static const struct
ltc2632_chip_info objects for each supported chip variant.

Suggested-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: David Marinovic <david.marinovic@pupin.rs>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agostaging: iio: ad9834: use sysfs_emit() and simplify show functions
Gabriel Rondon [Fri, 20 Mar 2026 22:24:24 +0000 (22:24 +0000)] 
staging: iio: ad9834: use sysfs_emit() and simplify show functions

Replace sprintf() with sysfs_emit() in sysfs attribute show functions.
sysfs_emit() is the preferred API for sysfs callbacks as it is aware
of the PAGE_SIZE buffer limit.

Also simplify the wavetype_available show functions by removing
the intermediate string variable and returning directly from each
branch.

Signed-off-by: Gabriel Rondon <grondon@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agostaging: iio: ad5933: use sysfs_emit() in show functions
Gabriel Rondon [Fri, 20 Mar 2026 22:24:23 +0000 (22:24 +0000)] 
staging: iio: ad5933: use sysfs_emit() in show functions

Replace sprintf() with sysfs_emit() in all sysfs attribute show
functions. sysfs_emit() is the preferred API for sysfs callbacks as
it is aware of the PAGE_SIZE buffer limit.

Also remove the unnecessary (int) cast in ad5933_show_frequency()
and use the correct format specifier %llu for the unsigned long long
freqreg variable.

Signed-off-by: Gabriel Rondon <grondon@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: vcnl4000: add support for regulators
Erikas Bitovtas [Fri, 20 Mar 2026 16:45:43 +0000 (18:45 +0200)] 
iio: light: vcnl4000: add support for regulators

Add supply, I2C and cathode voltage regulators to the sensor and enable
them. This keeps the sensor powered on even after its only supply shared
by another device shuts down.

Reported-by: Raymond Hackley <raymondhackley@protonmail.com>
Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Erikas Bitovtas <xerikasxx@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: vcnl4000: remove redundant check for proximity-near-level
Erikas Bitovtas [Fri, 20 Mar 2026 16:45:42 +0000 (18:45 +0200)] 
iio: light: vcnl4000: remove redundant check for proximity-near-level

The data->near_level variable is already assigned 0 during
devm_kzalloc(), therefore checking if the property is present and then
assigning it 0 is redundant. Remove the check for device tree property
and let it fail silently if it is missing or invalid.

Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Erikas Bitovtas <xerikasxx@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: vcnl4000: use variables for I2C client and device instances
Erikas Bitovtas [Fri, 20 Mar 2026 16:45:41 +0000 (18:45 +0200)] 
iio: light: vcnl4000: use variables for I2C client and device instances

After moving data->client and client->dev into variables of their own,
replace all instances of data->client and client->dev being used in
vcnl4200_init() and vcnl4000_probe() by the said variables to reduce
clutter.

Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Erikas Bitovtas <xerikasxx@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: vcnl4000: remove error messages for trigger and irq
Erikas Bitovtas [Fri, 20 Mar 2026 16:45:40 +0000 (18:45 +0200)] 
iio: light: vcnl4000: remove error messages for trigger and irq

The error code is available in the log after return. In our case,
attaching a triggered buffer can only fail if we are out of memory, as
no other buffer is being attached. Remove duplicate error messages to
reduce noise in dmesg.

Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Erikas Bitovtas <xerikasxx@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: vcnl4000: replace mutex_init() with devm_mutex_init()
Erikas Bitovtas [Fri, 20 Mar 2026 16:45:39 +0000 (18:45 +0200)] 
iio: light: vcnl4000: replace mutex_init() with devm_mutex_init()

Replace mutex_init() used across the driver with its device-managed
counterpart, so all assigned mutexes get destroyed.

Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Erikas Bitovtas <xerikasxx@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: vcnl4000: move power enablement from init to probe
Erikas Bitovtas [Fri, 20 Mar 2026 16:45:38 +0000 (18:45 +0200)] 
iio: light: vcnl4000: move power enablement from init to probe

Given both vcnl4000_init() and vcnl4200_init() end with
dev->chip_spec->set_power_state(), they can be called once from the
probe to enable the sensors. Move the set_power_state function from init
and call it after init function in probe.

Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Erikas Bitovtas <xerikasxx@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: vcnl4000: sort includes by their name
Erikas Bitovtas [Fri, 20 Mar 2026 16:45:37 +0000 (18:45 +0200)] 
iio: light: vcnl4000: sort includes by their name

Sort include headers by file name for better readability.

Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Erikas Bitovtas <xerikasxx@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agodt-bindings: iio: light: vcnl4000: add regulators
Erikas Bitovtas [Fri, 20 Mar 2026 16:45:36 +0000 (18:45 +0200)] 
dt-bindings: iio: light: vcnl4000: add regulators

These sensors can accept 2 supplies - one for the sensor and one for IR
LED [1]. Add supply properties for the sensor - 2 for the sensors and
one external, for their open drain interrupt line, to ensure the sensor
is powered on before proceeding with setup.

[1] https://www.vishay.com/docs/84274/vcnl4040.pdf

Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Erikas Bitovtas <xerikasxx@gmail.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: backend: use __free(fwnode_handle) for automatic cleanup
Antoniu Miclaus [Sat, 21 Mar 2026 10:01:51 +0000 (12:01 +0200)] 
iio: backend: use __free(fwnode_handle) for automatic cleanup

Convert __devm_iio_backend_fwnode_get() to use the __free(fwnode_handle)
cleanup attribute for the fwnode_back variable, removing the need for
manual fwnode_handle_put() calls. Move the declaration closer to its
first use, narrowing its scope.

No functional change.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Reviewed-by: Nuno Sá <nuno.sa@analog.com>
Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: accel: adxl372: add support for ADXL371
Antoniu Miclaus [Sat, 21 Mar 2026 10:04:59 +0000 (12:04 +0200)] 
iio: accel: adxl372: add support for ADXL371

Add support for the Analog Devices ADXL371, a +-200g 3-axis MEMS
accelerometer sharing the same register map as the ADXL372 but with
different ODR values (320/640/1280/2560/5120 Hz vs 400/800/1600/3200/
6400 Hz), different bandwidth values, and different timer scale
factors for activity/inactivity detection.

Due to a silicon anomaly (er001) causing FIFO data misalignment on
all current ADXL371 silicon, FIFO and triggered buffer support is
disabled for the ADXL371 - only direct mode reads are supported.

Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: accel: adxl372: factor out buffer and trigger setup
Antoniu Miclaus [Sat, 21 Mar 2026 10:04:58 +0000 (12:04 +0200)] 
iio: accel: adxl372: factor out buffer and trigger setup

Extract the triggered buffer, trigger allocation, and IRQ request
logic from adxl372_probe() into a dedicated adxl372_buffer_setup()
helper. This reduces the probe function complexity and prepares for
conditionally disabling buffer support on device variants with
known FIFO issues.

No functional change intended.

Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agodt-bindings: iio: accel: adi,adxl372: add ADXL371 compatible
Antoniu Miclaus [Sat, 21 Mar 2026 10:04:57 +0000 (12:04 +0200)] 
dt-bindings: iio: accel: adi,adxl372: add ADXL371 compatible

Add the adi,adxl371 compatible string to the ADXL372 binding. The
ADXL371 is a +-200g 3-axis MEMS accelerometer nearly identical to
the ADXL372 in register layout, differing only in ODR/bandwidth
values, timer scale factors, and a silicon anomaly affecting FIFO
operation.

Update the title and description to reflect both devices.

Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: accel: adxl372: introduce chip_info structure
Antoniu Miclaus [Sat, 21 Mar 2026 10:04:56 +0000 (12:04 +0200)] 
iio: accel: adxl372: introduce chip_info structure

Introduce a chip_info structure to parameterize device-specific
properties such as ODR/bandwidth frequency tables, activity/inactivity
timer scale factors, and the maximum ODR value. This refactors the
driver to use chip_info lookups instead of hardcoded values, preparing
the driver to support multiple device variants.

The sampling_frequency and filter_low_pass_3db_frequency available
attributes are switched from custom sysfs callbacks to read_avail()
based handling via info_mask_shared_by_type_available. This enforces
consistent formatting through the IIO framework and makes the values
accessible to in-kernel consumers.

The SPI/I2C probe functions are updated to pass a chip_info pointer
instead of a device name string.

No functional change intended.

Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoLinux 7.0-rc4 v7.0-rc4
Linus Torvalds [Sun, 15 Mar 2026 20:52:05 +0000 (13:52 -0700)] 
Linux 7.0-rc4

2 months agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Sun, 15 Mar 2026 20:15:39 +0000 (13:15 -0700)] 
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "The one core change is a re-roll of the tag allocation fix from the
  last pull request that uses the correct goto to unroll all the
  allocations. The remianing fixes are all small ones in drivers"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: hisi_sas: Fix NULL pointer exception during user_scan()
  scsi: qla2xxx: Completely fix fcport double free
  scsi: ufs: core: Fix SError in ufshcd_rtc_work() during UFS suspend
  scsi: core: Fix error handling for scsi_alloc_sdev()

2 months agoMerge tag 'probes-fixes-v7.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 15 Mar 2026 20:08:05 +0000 (13:08 -0700)] 
Merge tag 'probes-fixes-v7.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

Pull probes fixes from Masami Hiramatsu:

 - Avoid crash when rmmod/insmod after ftrace killed

   This fixes a kernel crash caused by kprobes on the symbol in a module
   which is unloaded after ftrace_kill() is called.

 - Remove unneeded warnings from __arm_kprobe_ftrace()

   Remove unneeded WARN messages which can be triggered if the kprobe is
   using ftrace and it fails to enable the ftrace. Since kprobes
   correctly handle such failure, we don't need to warn it.

* tag 'probes-fixes-v7.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  kprobes: Remove unneeded warnings from __arm_kprobe_ftrace()
  kprobes: avoid crash when rmmod/insmod after ftrace killed

2 months agoMerge tag 'bootconfig-fixes-v7.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 15 Mar 2026 19:50:05 +0000 (12:50 -0700)] 
Merge tag 'bootconfig-fixes-v7.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

Pull bootconfig fixes from Masami Hiramatsu:

 - fix off-by-one in xbc_verify_tree() unclosed brace error. This fixes
   a wrong error place in unclosed brace error message

 - check bounds before writing in __xbc_open_brace(). This fixes to
   check the array index before setting array, so that the bootconfig
   can support 16th-depth nested brace correctly

 - fix snprintf truncation check in xbc_node_compose_key_after(). This
   fixes to handle the return value of snprintf() correctly in case of
   the return value == size

 - Add bootconfig tests about braces Add test cases for checking error
   position about unclosed brace and ensuring supporting 16th depth
   nested braces correctly

* tag 'bootconfig-fixes-v7.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  bootconfig: Add bootconfig tests about braces
  lib/bootconfig: fix snprintf truncation check in xbc_node_compose_key_after()
  lib/bootconfig: check bounds before writing in __xbc_open_brace()
  lib/bootconfig: fix off-by-one in xbc_verify_tree() unclosed brace error

2 months agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Sun, 15 Mar 2026 19:22:10 +0000 (12:22 -0700)] 
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull kvm fixes from Paolo Bonzini:
 "Quite a large pull request, partly due to skipping last week and
  therefore having material from ~all submaintainers in this one. About
  a fourth of it is a new selftest, and a couple more changes are large
  in number of files touched (fixing a -Wflex-array-member-not-at-end
  compiler warning) or lines changed (reformatting of a table in the API
  documentation, thanks rST).

  But who am I kidding---it's a lot of commits and there are a lot of
  bugs being fixed here, some of them on the nastier side like the
  RISC-V ones.

  ARM:

   - Correctly handle deactivation of interrupts that were activated
     from LRs. Since EOIcount only denotes deactivation of interrupts
     that are not present in an LR, start EOIcount deactivation walk
     *after* the last irq that made it into an LR

   - Avoid calling into the stubs to probe for ICH_VTR_EL2.TDS when pKVM
     is already enabled -- not only thhis isn't possible (pKVM will
     reject the call), but it is also useless: this can only happen for
     a CPU that has already booted once, and the capability will not
     change

   - Fix a couple of low-severity bugs in our S2 fault handling path,
     affecting the recently introduced LS64 handling and the even more
     esoteric handling of hwpoison in a nested context

   - Address yet another syzkaller finding in the vgic initialisation,
     where we would end-up destroying an uninitialised vgic with nasty
     consequences

   - Address an annoying case of pKVM failing to boot when some of the
     memblock regions that the host is faulting in are not page-aligned

   - Inject some sanity in the NV stage-2 walker by checking the limits
     against the advertised PA size, and correctly report the resulting
     faults

  PPC:

   - Fix a PPC e500 build error due to a long-standing wart that was
     exposed by the recent conversion to kmalloc_obj(); rip out all the
     ugliness that led to the wart

  RISC-V:

   - Prevent speculative out-of-bounds access using array_index_nospec()
     in APLIC interrupt handling, ONE_REG regiser access, AIA CSR
     access, float register access, and PMU counter access

   - Fix potential use-after-free issues in kvm_riscv_gstage_get_leaf(),
     kvm_riscv_aia_aplic_has_attr(), and kvm_riscv_aia_imsic_has_attr()

   - Fix potential null pointer dereference in
     kvm_riscv_vcpu_aia_rmw_topei()

   - Fix off-by-one array access in SBI PMU

   - Skip THP support check during dirty logging

   - Fix error code returned for Smstateen and Ssaia ONE_REG interface

   - Check host Ssaia extension when creating AIA irqchip

  x86:

   - Fix cases where CPUID mitigation features were incorrectly marked
     as available whenever the kernel used scattered feature words for
     them

   - Validate _all_ GVAs, rather than just the first GVA, when
     processing a range of GVAs for Hyper-V's TLB flush hypercalls

   - Fix a brown paper bug in add_atomic_switch_msr()

   - Use hlist_for_each_entry_srcu() when traversing mask_notifier_list,
     to fix a lockdep warning; KVM doesn't hold RCU, just irq_srcu

   - Ensure AVIC VMCB fields are initialized if the VM has an in-kernel
     local APIC (and AVIC is enabled at the module level)

   - Update CR8 write interception when AVIC is (de)activated, to fix a
     bug where the guest can run in perpetuity with the CR8 intercept
     enabled

   - Add a quirk to skip the consistency check on FREEZE_IN_SMM, i.e. to
     allow L1 hypervisors to set FREEZE_IN_SMM. This reverts (by
     default) an unintentional tightening of userspace ABI in 6.17, and
     provides some amount of backwards compatibility with hypervisors
     who want to freeze PMCs on VM-Entry

   - Validate the VMCS/VMCB on return to a nested guest from SMM,
     because either userspace or the guest could stash invalid values in
     memory and trigger the processor's consistency checks

  Generic:

   - Remove a subtle pseudo-overlay of kvm_stats_desc, which, aside from
     being unnecessary and confusing, triggered compiler warnings due to
     -Wflex-array-member-not-at-end

   - Document that vcpu->mutex is take outside of kvm->slots_lock and
     kvm->slots_arch_lock, which is intentional and desirable despite
     being rather unintuitive

  Selftests:

   - Increase the maximum number of NUMA nodes in the guest_memfd
     selftest to 64 (from 8)"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (43 commits)
  KVM: selftests: Verify SEV+ guests can read and write EFER, CR0, CR4, and CR8
  Documentation: kvm: fix formatting of the quirks table
  KVM: x86: clarify leave_smm() return value
  selftests: kvm: add a test that VMX validates controls on RSM
  selftests: kvm: extract common functionality out of smm_test.c
  KVM: SVM: check validity of VMCB controls when returning from SMM
  KVM: VMX: check validity of VMCS controls when returning from SMM
  KVM: SVM: Set/clear CR8 write interception when AVIC is (de)activated
  KVM: SVM: Initialize AVIC VMCB fields if AVIC is enabled with in-kernel APIC
  KVM: x86: Introduce KVM_X86_QUIRK_VMCS12_ALLOW_FREEZE_IN_SMM
  KVM: x86: Fix SRCU list traversal in kvm_fire_mask_notifiers()
  KVM: VMX: Fix a wrong MSR update in add_atomic_switch_msr()
  KVM: x86: hyper-v: Validate all GVAs during PV TLB flush
  KVM: x86: synthesize CPUID bits only if CPU capability is set
  KVM: PPC: e500: Rip out "struct tlbe_ref"
  KVM: PPC: e500: Fix build error due to using kmalloc_obj() with wrong type
  KVM: selftests: Increase 'maxnode' for guest_memfd tests
  KVM: arm64: pkvm: Don't reprobe for ICH_VTR_EL2.TDS on CPU hotplug
  KVM: arm64: vgic: Pick EOIcount deactivations from AP-list tail
  KVM: arm64: Remove the redundant ISB in __kvm_at_s1e2()
  ...

2 months agoMerge tag 'powerpc-7.0-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc...
Linus Torvalds [Sun, 15 Mar 2026 18:36:11 +0000 (11:36 -0700)] 
Merge tag 'powerpc-7.0-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux

Pull powerpc fixes from Madhavan Srinivasan:

 - Fix KUAP warning in VMX usercopy path

 - Fix lockdep warning during PCI enumeration

 - Fix to move CMA reservations to arch_mm_preinit

 - Fix to check current->mm is alive before getting user callchain

Thanks to Aboorva Devarajan, Christophe Leroy (CS GROUP), Dan Horák,
Nicolin Chen, Nilay Shroff, Qiao Zhao, Ritesh Harjani (IBM), Saket Kumar
Bhaskar, Sayali Patil, Shrikanth Hegde, Venkat Rao Bagalkote, and Viktor
Malik.

* tag 'powerpc-7.0-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/iommu: fix lockdep warning during PCI enumeration
  powerpc/selftests/copyloops: extend selftest to exercise __copy_tofrom_user_power7_vmx
  powerpc: fix KUAP warning in VMX usercopy path
  powerpc, perf: Check that current->mm is alive before getting user callchain
  powerpc/mem: Move CMA reservations to arch_mm_preinit

2 months agoMerge tag 'x86-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 15 Mar 2026 18:26:36 +0000 (11:26 -0700)] 
Merge tag 'x86-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 fix from Ingo Molnar:
 "Work around S2RAM hang if the firmware unexpectedly re-enables the
  x2apic hardware while it was disabled by the kernel.

  Force-disable it again and issue a warning into the syslog"

* tag 'x86-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/apic: Disable x2apic on resume if the kernel expects so

2 months agoMerge tag 'timers-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 15 Mar 2026 18:14:09 +0000 (11:14 -0700)] 
Merge tag 'timers-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull timer fix from Ingo Molnar:
 "Fix function tracer recursion bug by marking jiffies_64_to_clock_t()
  notrace"

* tag 'timers-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  time/jiffies: Mark jiffies_64_to_clock_t() notrace

2 months agoMerge tag 'sched-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 15 Mar 2026 17:49:47 +0000 (10:49 -0700)] 
Merge tag 'sched-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull scheduler fixes from Ingo Molnar:
 "More MM-CID fixes, mostly fixing hangs/races:

   - Fix CID hangs due to a race between concurrent forks

   - Fix vfork()/CLONE_VM MMCID bug causing hangs

   - Remove pointless preemption guard

   - Fix CID task list walk performance regression on large systems
     by removing the known-flaky and slow counting logic using
     for_each_process_thread() in mm_cid_*fixup_tasks_to_cpus(), and
     implementing a simple sched_mm_cid::node list instead"

* tag 'sched-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/mmcid: Avoid full tasklist walks
  sched/mmcid: Remove pointless preempt guard
  sched/mmcid: Handle vfork()/CLONE_VM correctly
  sched/mmcid: Prevent CID stalls due to concurrent forks

2 months agoMerge tag 'objtool-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 15 Mar 2026 17:36:01 +0000 (10:36 -0700)] 
Merge tag 'objtool-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull objtool fixes from Ingo Molnar:

 - Fix cross-build bug by using HOSTCFLAGS for HAVE_XXHASH test

 - Fix klp bug by fixing detection of corrupt static branch/call entries

 - Handle unsupported pr_debug() usage more gracefully

 - Fix hypothetical klp bug by avoiding NULL pointer dereference when
   printing code symbol name

 - Fix data alignment bug in elf_add_data() causing mangled strings

 - Fix confusing ERROR_INSN() error message

 - Handle unexpected Clang RSP musical chairs causing false positive
   warnings

 - Fix another objtool stack overflow in validate_branch()

* tag 'objtool-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  objtool: Fix another stack overflow in validate_branch()
  objtool: Handle Clang RSP musical chairs
  objtool: Fix ERROR_INSN() error message
  objtool: Fix data alignment in elf_add_data()
  objtool: Use HOSTCFLAGS for HAVE_XXHASH test
  objtool/klp: Avoid NULL pointer dereference when printing code symbol name
  objtool/klp: Disable unsupported pr_debug() usage
  objtool/klp: Fix detection of corrupt static branch/call entries

2 months agoMerge tag 'irq-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 15 Mar 2026 17:32:57 +0000 (10:32 -0700)] 
Merge tag 'irq-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull irq fixes from Ingo Molnar:
 "Two fixes for the riscv-aplic irqchip driver:

   - Fix probing dependency bug on probing failure

   - Fix double register_syscore() bug"

* tag 'irq-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/riscv-aplic: Register syscore operations only once
  irqchip/riscv-aplic: Do not clear ACPI dependencies on probe failure

2 months agoiio: adc: mt6359-auxadc: Fix comma spacing
Giorgi Tchankvetadze [Tue, 10 Mar 2026 14:38:10 +0000 (18:38 +0400)] 
iio: adc: mt6359-auxadc: Fix comma spacing

Fix incorrect whitespace around comma on line 325 to comply with
kernel coding style. This silences checkpatch errors "ERROR: space
prohibited before that ','" and "ERROR: space required after that ','".
No functional change.

Signed-off-by: Giorgi Tchankvetadze <giorgitchankvetadze1997@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoMerge tag 'i3c/fixes-for-7.0' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c...
Linus Torvalds [Sat, 14 Mar 2026 23:25:10 +0000 (16:25 -0700)] 
Merge tag 'i3c/fixes-for-7.0' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux

Pull i3c fixes from Alexandre Belloni:
 "This introduces the I3C_OR_I2C symbol which is not a fix per se but is
  affecting multiple subsystems so it is included to ease
  synchronization.

  Apart from that, Adrian is mostly fixing the mipi-i3c-hci driver DMA
  handling, and I took the opportunity to add two fixes for the dw-i3c
  driver.

  Subsystem:
   - simplify combined i3c/i2c dependencies

  Drivers:
   - dw: handle 2C properly, fix possible race condition
   - mipi-i3c-hci: many DMA related fixes"

* tag 'i3c/fixes-for-7.0' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux:
  i3c: dw-i3c-master: Set SIR_REJECT in DAT on device attach and reattach
  i3c: master: dw-i3c: Fix missing of_node for virtual I2C adapter
  i3c: mipi-i3c-hci: Fallback to software reset when bus disable fails
  i3c: mipi-i3c-hci: Fix handling of shared IRQs during early initialization
  i3c: mipi-i3c-hci: Fix race in DMA error handling in interrupt context
  i3c: mipi-i3c-hci: Consolidate common xfer processing logic
  i3c: mipi-i3c-hci: Restart DMA ring correctly after dequeue abort
  i3c: mipi-i3c-hci: Add missing TID field to no-op command descriptor
  i3c: mipi-i3c-hci: Correct RING_CTRL_ABORT handling in DMA dequeue
  i3c: mipi-i3c-hci: Fix race between DMA ring dequeue and interrupt handler
  i3c: mipi-i3c-hci: Fix race in DMA ring dequeue
  i3c: mipi-i3c-hci: Fix race in DMA ring enqueue for parallel xfers
  i3c: mipi-i3c-hci: Consolidate spinlocks
  i3c: mipi-i3c-hci: Factor out DMA mapping from queuing path
  i3c: mipi-i3c-hci: Fix Hot-Join NACK
  i3c: mipi-i3c-hci: Use ETIMEDOUT instead of ETIME for timeout errors
  i3c: simplify combined i3c/i2c dependencies

2 months agoMerge tag 'i2c-for-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa...
Linus Torvalds [Sat, 14 Mar 2026 23:15:49 +0000 (16:15 -0700)] 
Merge tag 'i2c-for-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux

Pull i2c fix from Wolfram Sang:
 "Designware DT binding maintainer update"

* tag 'i2c-for-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  dt-bindings: i2c: dw: Update maintainer

2 months agoMerge tag 'rust-fixes-7.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda...
Linus Torvalds [Sat, 14 Mar 2026 19:35:16 +0000 (12:35 -0700)] 
Merge tag 'rust-fixes-7.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux

Pull Rust fixes from Miguel Ojeda:
 "Toolchain and infrastructure:

   - Remap paths to avoid absolute ones starting with the upcoming Rust
     1.95.0 release. This improves build reproducibility, avoids leaking
     the exact path and avoids having the same path appear in two forms

     The approach here avoids remapping debug information as well, in
     order to avoid breaking tools that used the paths to access source
     files, which was the previous attempt that needed to be reverted

   - Allow 'unused_features' lint for the upcoming Rust 1.96.0 release.
     While well-intentioned, we do not benefit much from the new lint

   - Emit dependency information into '$(depfile)' directly to avoid a
     temporary '.d' file (it was an old approach)

  'kernel' crate:

   - 'str' module: fix warning under '!CONFIG_BLOCK' by making
     'NullTerminatedFormatter' public

   - 'cpufreq' module: suppress false positive Clippy warning

  'pin-init' crate:

   - Remove '#[disable_initialized_field_access]' attribute which was
     unsound. This means removing the support for structs with unaligned
     fields (through the 'repr(packed)' attribute), for now

     And document the load-bearing fact of field accessors (i.e. that
     they are required for soundness)

   - Replace shadowed return token by 'unsafe'-to-create token in order
     to remain sound in the face of the likely upcoming Type Alias Impl
     Trait (TAIT) and the next trait solver in upstream Rust"

* tag 'rust-fixes-7.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux:
  rust: kbuild: allow `unused_features`
  rust: cpufreq: suppress clippy::double_parens in Policy doctest
  rust: pin-init: replace shadowed return token by `unsafe`-to-create token
  rust: pin-init: internal: init: document load-bearing fact of field accessors
  rust: pin-init: internal: init: remove `#[disable_initialized_field_access]`
  rust: build: remap path to avoid absolute path
  rust: kbuild: emit dep-info into $(depfile) directly
  rust: str: make NullTerminatedFormatter public

2 months agoMerge tag 'usb-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Linus Torvalds [Sat, 14 Mar 2026 16:43:12 +0000 (09:43 -0700)] 
Merge tag 'usb-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here is a large chunk of USB driver fixes for 7.0-rc4. Included in
  here are:

   - usb gadget reverts due to reported issues, and then a follow-on fix
     to hopefully resolve the reported overall problem

   - xhci driver fixes

   - dwc3 driver fixes

   - usb core "killable" bulk message api addition to fix a usbtmc
     driver bug where userspace could hang the driver for forever

   - small USB driver fixes for reported issues

   - new usb device quirks

  All except the last USB device quirk change have been in linux-next
  with no reported issues. That one came in too late, and is 'obviously
  correct' :)"

* tag 'usb-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (35 commits)
  USB: ezcap401 needs USB_QUIRK_NO_BOS to function on 10gbs usb speed
  usb: roles: get usb role switch from parent only for usb-b-connector
  Revert "tcpm: allow looking for role_sw device in the main node"
  usb: gadget: f_ncm: Fix net_device lifecycle with device_move
  Revert "usb: gadget: u_ether: add gether_opts for config caching"
  Revert "usb: gadget: u_ether: use <linux/hex.h> header file"
  Revert "usb: gadget: u_ether: Add auto-cleanup helper for freeing net_device"
  Revert "usb: gadget: f_ncm: align net_device lifecycle with bind/unbind"
  Revert "usb: legacy: ncm: Fix NPE in gncm_bind"
  Revert "usb: gadget: f_ncm: Fix atomic context locking issue"
  usb: typec: altmode/displayport: set displayport signaling rate in configure message
  usb: dwc3: pci: add support for the Intel Nova Lake -H
  usb/core/quirks: Add Huawei ME906S-device to wakeup quirk
  usb: gadget: uvc: fix interval_duration calculation
  xhci: Fix NULL pointer dereference when reading portli debugfs files
  usb: xhci: Prevent interrupt storm on host controller error (HCE)
  usb: xhci: Fix memory leak in xhci_disable_slot()
  usb: class: cdc-wdm: fix reordering issue in read code path
  usb: renesas_usbhs: fix use-after-free in ISR during device removal
  usb: cdc-acm: Restore CAP_BRK functionnality to CH343
  ...

2 months agoMerge tag 'char-misc-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Sat, 14 Mar 2026 16:38:49 +0000 (09:38 -0700)] 
Merge tag 'char-misc-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc

Pull char / misc / IIO driver fixes from Greg KH:
 "Here are some char/misc/iio/binder fixes for 7.0-rc4. Nothing major in
  here, just the usual:

   - lots of iio driver fixes for reported issues

   - rust binder fixes for problems found

   - gpib driver binding to the wrong device fix

   - firmware driver fix

  All of these have been in linux-next with no reported issues"

* tag 'char-misc-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (28 commits)
  gpib: lpvo_usb: fix unintended binding of FTDI 8U232AM devices
  firmware: stratix10-svc: Add Multi SVC clients support
  rust_binder: use lock_vma_under_rcu() in use_page_slow()
  rust_binder: call set_notification_done() without proc lock
  rust_binder: avoid reading the written value in offsets array
  rust_binder: check ownership before using vma
  rust_binder: fix oneway spam detection
  firmware: stratix10-rsu: Fix NULL pointer dereference when RSU is disabled
  iio: imu: adis: Fix NULL pointer dereference in adis_init
  iio: imu: inv_icm45600: fix regulator put warning when probe fails
  iio: buffer: Fix wait_queue not being removed
  iio: gyro: mpu3050-core: fix pm_runtime error handling
  iio: gyro: mpu3050-i2c: fix pm_runtime error handling
  iio: adc: ad7768-1: Fix ERR_PTR dereference in ad7768_fill_scale_tbl
  iio: chemical: sps30_serial: fix buffer size in sps30_serial_read_meas()
  iio: chemical: sps30_i2c: fix buffer size in sps30_i2c_read_meas()
  iio: magnetometer: tlv493d: remove erroneous shift in X-axis data
  iio: proximity: hx9023s: Protect against division by zero in set_samp_freq
  iio: proximity: hx9023s: fix assignment order for __counted_by
  iio: chemical: bme680: Fix measurement wait duration calculation
  ...

2 months agoMerge tag 'staging-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Sat, 14 Mar 2026 16:33:58 +0000 (09:33 -0700)] 
Merge tag 'staging-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging

Pull staging driver fixes from Greg KH:
 "Here are three small staging driver fixes for 7.0-rc4 that resolve
  some reported problems. They are:

   - two rtl8723bs data validation bugfixes

   - sm750fb removal path bugfix

  All of these have been in linux-next for many weeks with no reported
  issues"

* tag 'staging-7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: rtl8723bs: fix potential out-of-bounds read in rtw_restruct_wmm_ie
  staging: rtl8723bs: properly validate the data in rtw_get_ie_ex()
  staging: sm750fb: add missing pci_release_region on error and removal

2 months agoiio: adc: ti-ads7950: remove chip_info[]
David Lechner [Sat, 7 Mar 2026 21:54:37 +0000 (15:54 -0600)] 
iio: adc: ti-ads7950: remove chip_info[]

Remove the chip_info[] array and related enum used for looking up chip-
specific information. Instead, use individual structs directly in the
module device tables.

Also update to use spi_get_device_match_data() in case the devicetree
table is ever used instead of the SPI device ID table.

Signed-off-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: palmas_gpadc: Replace leading space indentation with tabs
Giorgi Tchankvetadze [Sat, 7 Mar 2026 13:02:14 +0000 (17:02 +0400)] 
iio: adc: palmas_gpadc: Replace leading space indentation with tabs

Fix lines starting with spaces instead of tabs to comply with the
kernel coding style.

No functional change.

Signed-off-by: Giorgi Tchankvetadze <giorgitchankvetadze1997@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: accel: adxl372: convert to guard(mutex)
Rajveer Chaudhari [Sat, 7 Mar 2026 11:49:12 +0000 (17:19 +0530)] 
iio: accel: adxl372: convert to guard(mutex)

Replace manual mutex_lock/mutex_unlock pair with guard(mutex) in
adxl372_write_threshold_value(). This ensures the mutex is released
on all return paths and allows returning directly without a goto label.

Signed-off-by: Rajveer Chaudhari <rajveer.chaudhari.linux@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: accel: adxl313: convert to guard(mutex)
Rajveer Chaudhari [Sat, 7 Mar 2026 11:49:11 +0000 (17:19 +0530)] 
iio: accel: adxl313: convert to guard(mutex)

Replace manual mutex_lock/mutex_unlock pair with guard(mutex) in
adxl313_read_axis(). This ensures the mutex is released on all
return paths and allows returning directly without a goto label.

Signed-off-by: Rajveer Chaudhari <rajveer.chaudhari.linux@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: buffer: hw-consumer: remove redundant scan_mask flexible array
Nuno Sá [Tue, 3 Mar 2026 11:50:44 +0000 (11:50 +0000)] 
iio: buffer: hw-consumer: remove redundant scan_mask flexible array

The scan_mask flexible array member in hw_consumer_buffer duplicates
the scan_mask pointer already present in struct iio_buffer. Remove it
and allocate the bitmap directly with bitmap_zalloc(), assigning it to
buf->buffer.scan_mask. This simplifies the code and there's no need for
the flex array allocation.

Signed-off-by: Nuno Sá <nuno.sa@analog.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: meson: no devm for nvmem_cell_get
Rosen Penev [Wed, 4 Mar 2026 04:17:16 +0000 (20:17 -0800)] 
iio: adc: meson: no devm for nvmem_cell_get

There is no reason to extend the lifetime of this post removal of the
driver when it's only needed in one spot.

Moved tsc_regmap assignment to avoid two nvmem_cell_put calls.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: at91-sama5d2_adc: no devm for nvmem_cell_get
Rosen Penev [Wed, 4 Mar 2026 04:18:43 +0000 (20:18 -0800)] 
iio: adc: at91-sama5d2_adc: no devm for nvmem_cell_get

There is absolutely no reason to pospone cleanup of this post driver
removal. Just do it immediately.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Reviewed-by: Nuno Sá <nuno.sa@analog.com>
Reviewed-by: Eugen Hristev <eugen.hristev@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: acpi-als: Convert ACPI driver to a platform one
Rafael J. Wysocki [Wed, 4 Mar 2026 18:33:14 +0000 (19:33 +0100)] 
iio: light: acpi-als: Convert ACPI driver to a platform one

In all cases in which a struct acpi_driver is used for binding a driver
to an ACPI device object, a corresponding platform device is created by
the ACPI core and that device is regarded as a proper representation of
underlying hardware.  Accordingly, a struct platform_driver should be
used by driver code to bind to that device.  There are multiple reasons
why drivers should not bind directly to ACPI device objects [1].

Overall, it is better to bind drivers to platform devices than to their
ACPI companions, so convert the ACPI ambient light sensor driver to a
platform one.

After this change, the subordinate IIO device will be registered under
the platform device used for driver binding instead of its ACPI
companion.

While this is not expected to alter functionality, it changes sysfs
layout and so it will be visible to user space.

Link: https://lore.kernel.org/all/2396510.ElGaqSPkdT@rafael.j.wysocki/
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: light: acpi-als: Register ACPI notify handler directly
Rafael J. Wysocki [Wed, 4 Mar 2026 18:32:26 +0000 (19:32 +0100)] 
iio: light: acpi-als: Register ACPI notify handler directly

To facilitate subsequent conversion of the driver to a platform one,
make it install an ACPI notify handler directly instead of using
a .notify() callback in struct acpi_driver.

No intentional functional impact.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: frequency: admv1013: fix NULL pointer dereference on str
Antoniu Miclaus [Thu, 5 Mar 2026 09:14:48 +0000 (11:14 +0200)] 
iio: frequency: admv1013: fix NULL pointer dereference on str

When device_property_read_string() fails, str is left uninitialized
but the code falls through to strcmp(str, ...), dereferencing a garbage
pointer. Replace manual read/strcmp with
device_property_match_property_string() and consolidate the SE mode
enums into a single sequential enum, mapping to hardware register
values via a switch consistent with other bitfields in the driver.

Several cleanup patches have been applied to this driver recently so
this will need a manual backport.

Fixes: da35a7b526d9 ("iio: frequency: admv1013: add support for ADMV1013")
Reviewed-by: Nuno Sá <nuno.sa@analog.com>
Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoiio: adc: max1363: Reformat enum and array initializers
Giorgi Tchankvetadze [Sat, 7 Mar 2026 07:15:56 +0000 (11:15 +0400)] 
iio: adc: max1363: Reformat enum and array initializers

Reformat the device enum so each entry is on its own line and add a
trailing comma to the final enumerator. Also reformat the nearby monitor
speeds array for consistency.

No functional change.

Signed-off-by: Giorgi Tchankvetadze <giorgitchankvetadze1997@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2 months agoMerge tag 'drm-fixes-2026-03-14' of https://gitlab.freedesktop.org/drm/kernel
Linus Torvalds [Fri, 13 Mar 2026 22:38:55 +0000 (15:38 -0700)] 
Merge tag 'drm-fixes-2026-03-14' of https://gitlab.freedesktop.org/drm/kernel

Pull drm fixes from Dave Airlie:
 "The weekly drm fixes. This is mostly msm fixes across the functions,
  with amdgpu and i915. It also has a core rust fix and changes in
  nova-core to take advantage of it, and otherwise just has some minor
  driver fixes, and marks loongsoon as orphaned.

  rust:
   - Fix safety issue in dma_read! and dma_write!

  nova-core:
   - Fix UB in DmaGspMem pointer accessors
   - Fix stack overflow in GSP memory allocation

  loongsoon:
   - mark drm driver as unmaintained

  msm:
   - Core:
      - Adjusted msm_iommu_pagetable_prealloc_allocate() allocation type
   - DPU:
      - Fixed blue screens on Hamoa laptops by reverting the LM
        reservation
      - Fixed the size of the LM block on several platforms
      - Dropped usage of %pK (again)
      - Fixed smatch warning on SSPP v13+ code
      - Fixed INTF_6 interrupts on Lemans
   - DSI:
      - Fixed DSI PHY revision on Kaanapali
      - Fixed pixel clock calculation for the bonded DSI mode panels
        with compression enabled
   - DT bindings:
      - Fixed DisplayPort description on Glymur
      - Fixed model name in SM8750 MDSS schema
   - GPU:
      - Added MODULE_DEVICE_TABLE to the GPU driver
      - Fix bogus protect error on X2-85
      - Fix dma_free_attrs() buffer size
      - Gen8 UBWC fix for Glymur

  i915:
   - Avoid hang when configuring VRR [icl]
   - Fix sg_table overflow with >4GB folios
   - Fix PSR Selective Update handling
   - Fix eDP ALPM read-out sequence

  amdgpu:
   - SMU13 fix
   - SMU14 fix
   - Fixes for bringup hw testing
   - Kerneldoc fix
   - GC12 idle power fix for compute workloads
   - DCCG fixes

  amdkfd:
   - Fix missing BO unreserve in an error path

  ivpu:
   - drop unnecessary bootparams register setting

  amdxdna:
   - fix runtime/suspend resume deadlock

  bridge:
   - ti-sn65dsi83: fix DSI rounding and dual LVDS

  gud:
   - fix NULL crtc dereference on display disable"

* tag 'drm-fixes-2026-03-14' of https://gitlab.freedesktop.org/drm/kernel: (44 commits)
  drm/amd: Set num IP blocks to 0 if discovery fails
  drm/amdkfd: Unreserve bo if queue update failed
  drm/amd/display: Check for S0i3 to be done before DCCG init on DCN21
  drm/amd/display: Add missing DCCG register entries for DCN20-DCN316
  gpu: nova-core: gsp: fix UB in DmaGspMem pointer accessors
  drm/loongson: Mark driver as orphaned
  accel/amdxdna: Fix runtime suspend deadlock when there is pending job
  gpu: nova-core: fix stack overflow in GSP memory allocation
  accel/ivpu: Remove boot params address setting via MMIO register
  drm/i915/dp: Read ALPM caps after DPCD init
  drm/i915/psr: Write DSC parameters on Selective Update in ET mode
  drm/i915/dsc: Add helper for writing DSC Selective Update ET parameters
  drm/i915/dsc: Add Selective Update register definitions
  drm/i915/psr: Repeat Selective Update area alignment
  drm/i915: Fix potential overflow of shmem scatterlist length
  drm/i915/vrr: Configure VRR timings after enabling TRANS_DDI_FUNC_CTL
  drm/bridge: ti-sn65dsi83: halve horizontal syncs for dual LVDS output
  drm/bridge: ti-sn65dsi83: fix CHA_DSI_CLK_RANGE rounding
  drm/gud: fix NULL crtc dereference on display disable
  drm/sitronix/st7586: fix bad pixel data due to byte swap
  ...

2 months agoMerge tag 'wq-for-7.0-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 13 Mar 2026 22:11:05 +0000 (15:11 -0700)] 
Merge tag 'wq-for-7.0-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq

Pull workqueue fixes from Tejun Heo:

 - Improve workqueue stall diagnostics: dump all busy workers (not just
   running ones), show wall-clock duration of in-flight work items, and
   add a sample module for reproducing stalls

 - Fix POOL_BH vs WQ_BH flag namespace mismatch in pr_cont_worker_id()

 - Rename pool->watchdog_ts to pool->last_progress_ts and related
   functions for clarity

* tag 'wq-for-7.0-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
  workqueue: Rename show_cpu_pool{s,}_hog{s,}() to reflect broadened scope
  workqueue: Add stall detector sample module
  workqueue: Show all busy workers in stall diagnostics
  workqueue: Show in-flight work item duration in stall diagnostics
  workqueue: Rename pool->watchdog_ts to pool->last_progress_ts
  workqueue: Use POOL_BH instead of WQ_BH when checking pool flags

2 months agoMerge tag 'cgroup-for-7.0-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 13 Mar 2026 22:06:31 +0000 (15:06 -0700)] 
Merge tag 'cgroup-for-7.0-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup

Pull cgroup fixes from Tejun Heo:

 - Hide PF_EXITING tasks from cgroup.procs to avoid exposing dead tasks
   that haven't been removed yet, fixing a systemd timeout issue on
   PREEMPT_RT

 - Call rebuild_sched_domains() directly in CPU hotplug instead of
   deferring to a workqueue, fixing a race where online/offline CPUs
   could briefly appear in stale sched domains

* tag 'cgroup-for-7.0-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cgroup: Don't expose dead tasks in cgroup
  cgroup/cpuset: Call rebuild_sched_domains() directly in hotplug