Jonathan writes:

1st set of IIO/counter device support, features and cleanup in the 5.13 cycle

Big set in here from Alexandru Ardelean enabling multiple buffer support.
This includes providing a new directory per buffer that combines
what was previously in buffer/ and scan_elements/. Old interfaces still
in place for compatiblity.

Note immuatable branch for scmi patches to allow for some significant
rework going on in that subsystem.  Merge required updating to reflect
some changes in IIO.

Late rebase to fix some wrong fixes tags due to some earlier rebases
made necessary by messing up the immutable branch.

IIO New Device Support
* adi,ad5686
  - Add info to support AD5673R and AD5677R
* bosch,bmi088
  - New driver supporting this accelerometer + gyroscope
* cros_ec_mkbp
  - New driver for this proximity sensor that exposes a 'front'
    sensor. Very simple switch like device, but driver allows it
    to share interface with more sophisticated proximity sensors.
* iio_scmi
  - New driver to support ARM SCMI protocol to expose underlying
    accelerometers and gyroscopes via this firmware interface.
* st,st_magn
  - Add ID for IISMDC magnetometer.
* ti,ads131e0
  - New driver supporting ads131e04, ads131e06 and ads131e08 24 bit ADCs

Counter New Device Support
* IRQ or GPIO based counter
  - New driver for a conceptually simple counter that uses interrupts
    to perform the count.

Features
* core
  - Dual buffer supprt including:
     Various helpers to centralize handling of bufferer related elements.
     Document existing and new IOCTLs
     Register the IIO chrdev only if it can actually be used for anything.
     Rework attribute group creation in the core (lots of patches)
     Merge buffer/ and scan_elements/ entries into one list + maintain
     backwards compatible set.
     Introduce the internal logic and IOCTL to allow multiple buffers
     + access to an anon FD per buffer to actually read from it.
     Tidy up tools/iio/iio_generic_buffer and switch to new interfaces.
     Update ABI docs.
     A few follow up fixes, unsuprising as this was a huge bit of rework.
  - Move common case setting of trig->parent to the core.
  - Provide an iio_read_channel_processed_scale() to avoid loss of
    precision from iio_read_channel_processed() then applying integer
    scale. Use it in ntc_thermistor driver in hwmon.
  - Allow drivers to specify labels from elsewhere than DT. Use it for
    bmc150 and kxcjk-1013 labels related to position on 2 in one tablets.
  - Document label usage for proximity and accelerometer sensors.
  - Some local variable renames for consistency
tools
  - Add -a parameter to iio_event_monitor to allow autoenabling of events.
* acpi_als
  - Add trigger support for devices that don't support notification method.
* adi,ad7124
  - Allow more than 8 channels.  This is a complex little device, but is
    capable of supporting up to 16 channels if the share certain
    configuration settings.
* hrtimer-trigger
  - Support sampling frequency below 1Hz.
* mediatek,mt8195-auxadc
  - Add compatible to binding docs (always also includes mt8173)
* st,stm32-adc
  - Enable timetamps when not using DMA.
* vishay,vcnl3020
  - Sampling frequency control.

Cleanup and minor fixes:
* treewide
  - Use some getter and setter functions instead of opencoding.
  - Set of fixes for pointless casts in various drivers.
  - Avoid wrong kernel-doc marking on comment blocks.
  - Fix various other minor kernel-doc issues shown by W=1
* core
  - Use a signed temporary for IIO_VAL_FRACTIONAL_LOG2 to avoid odd casts.
  - Fix IIO_VAL_FRACTIONAL_LOG2 for values between -1.0 and 0.0
  - Add unit tests for iio_format_value()
* docs
  - Fix formatting/typos in iio_configfs.rst and buffers.rst
  - Add documentation of index in buffers.rst
  - Fix scan element description
  - Avoid some issues with HTML generation from ABI docs by moving
    duplicated defintions to more generic files.
  - Drop reference to long dead mailing list.
* 104-quad
  - Remove left over deprecated IIO counter ABI.
* adi,adi-axi-adc
  - Fix wrong bit of docs.
* adi,ad5791
  - Typos
* adi,ad9834
   - Switch to device managed functions in probe.
* adi,adis*
  - Add and use helpers for locking to reduced duplication.
* adi,adis16480
  - Fix calculation of sampling frequency when using pulse per second input.
* adi,adis16475
  - Calculate the IMU scaled internal sampling rate and runtime depending
    on sysfs based configuration rather than getting from DT. Drop now
    unnecessary property from DT bindings doc.
* cros_ec
   - Fix result of a series of recent changes that means extended buffer
     attributes turn up in the wrong place. Too complex to revert the
     various patches unfortunately so this is a bit messy.
* fsl,mma3452
  - Indentation cleanup.
* hid-sensors
  - Size of storage needs to increase for some parts when using quaternions.
  - Move the get sensistivity attribute to hid-sensors-common to reduce
    duplication.  Enable it for more device types.
  - Correctly handle relative sensitivity if reported that way including
    documenting the new ABI.
* maxim,max517
  - Use device managed functions in probe.
* mediatek,mt6360-adc
  - Use asm/unaligned.h instead of directly including
    unaligned/be_byteshift.h
* novuton,npcm-adc
  - Local lock instead of missusing mlock.
* semtech,sx9500
  - Typos
* st,sensor
  - typo fix
* st,spear-adc
  - Local lock instead of missusing mlock.
* st,stm32-adc
  - Long standing HAS_IOMEM dependency fix.
* st,stm32-counter
  - Remove left over deprecated IIO counter ABI.
* ti,palmas-adc
  - Local lock instead of missusing mlock.
* ti,tmp007
  - Switch to device managed functions in probe.

Other
* MAINTAINERS
  - Move Peter Meerwald-Stadler to Credits at his request

* tag 'iio-for-5.13a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio: (119 commits)
  iio: acpi_als: Add trigger support
  iio: acpi_als: Add local variable dev in probe
  iio: acpi_als: Add timestamp channel
  iio: adc: ad7292: Modify the bool initialization assignment
  iio: cros: unify hw fifo attributes without API changes
  iio: kfifo: add devm_iio_triggered_buffer_setup_ext variant
  iio: event_monitor: Enable events before monitoring
  dt-bindings: iio: adc: Add compatible for Mediatek MT8195
  iio:magnetometer: Add Support for ST IIS2MDC
  dt-bindings: iio: st,st-sensors add IIS2MDC.
  staging: iio: ad9832: kernel-doc fixes
  iio:dac:max517.c: Use devm_iio_device_register()
  iio:cros_ec_sensors: Fix a wrong function name in kernel doc.
  iio: buffer: kfifo_buf: kernel-doc, typo in function name.
  iio: accel: sca3000: kernel-doc fixes. Missing - and wrong function names.
  iio: adc: adi-axi-adc: Drop false marking for kernel-doc
  iio: adc: cpcap-adc: kernel-doc fix - that should be _ in structure name
  iio: dac: ad5504: fix wrong part number in kernel-doc structure name.
  iio: dac: ad5770r: kernel-doc fix case of letter R wrong in structure name
  iio: adc: ti-adc084s021: kernel-doc fixes, missing function names
  ...

Add simple IRQ or GPIO base counter. This device is used to measure
rotation speed of some agricultural devices, so no high frequency on the
counter pin is expected.

The maximal measurement frequency depends on the CPU and system load. On
the idle iMX6S I was able to measure up to 20kHz without count drops.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Reviewed-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Link: https://lore.kernel.org/r/20210301080401.22190-3-o.rempel@pengutronix.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

The IIO counter driver has been superseded by the Counter subsystem as
discussed in [1]. This patch removes the IIO counter ABI from the
104-QUAD-8 driver.

[1] https://lore.kernel.org/lkml/20210119104105.000010df@Huawei.com/

Cc: Syed Nayyar Waris <syednwaris@gmail.com>
Signed-off-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Link: https://lore.kernel.org/r/98a39983d5df761c058a469d1346fd8ffdef8516.1611973018.git.vilhelm.gray@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Currently, the STM32 LP Timer counter driver registers into both IIO and
counter subsystems, which is redundant.

Remove the IIO counter ABI and IIO registration from the STM32 LP Timer
counter driver since it's been superseded by the Counter subsystem
as discussed in [1].

Keep only the counter subsystem related part.
Move a part of the ABI documentation into a driver comment.

This also removes a duplicate ABI warning
$ scripts/get_abi.pl validate
...
/sys/bus/iio/devices/iio:deviceX/in_count0_preset is defined 2 times:
  ./Documentation/ABI/testing/sysfs-bus-iio-timer-stm32:100
  ./Documentation/ABI/testing/sysfs-bus-iio-lptimer-stm32:0

[1] https://lkml.org/lkml/2021/1/19/347

Acked-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@foss.st.com>
Link: https://lore.kernel.org/r/1611926542-2490-1-git-send-email-fabrice.gasnier@foss.st.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Ceiling value may be miss-aligned with what's actually configured into the
ARR register. This is seen after probe as currently the ARR value is zero,
whereas ceiling value is set to the maximum. So:
- reading ceiling reports zero
- in case the counter gets enabled without any prior configuration,
  it won't count.
- in case the function gets set by the user 1st, (priv->ceiling) is used.

Fix it by getting rid of the cached "priv->ceiling" variable. Rather use
the ARR register value directly by using regmap read or write when needed.
There should be no drawback on performance as priv->ceiling isn't used in
performance critical path.
There's also no point in writing ARR while setting function (sms), so
it can be safely removed.

Fixes: ad29937e206f ("counter: Add STM32 Timer quadrature encoder")
Suggested-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@foss.st.com>
Acked-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Cc: <Stable@vger.kernel.org>
Link: https://lore.kernel.org/r/1614793789-10346-1-git-send-email-fabrice.gasnier@foss.st.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

The ceiling value isn't checked before writing it into registers. The user
could write a value higher than the counter resolution (e.g. 16 or 32 bits
indicated by max_arr). This makes most significant bits to be truncated.
Fix it by checking the max_arr to report a range error [1] to the user.

[1] https://lkml.org/lkml/2021/2/12/358

Fixes: ad29937e206f ("counter: Add STM32 Timer quadrature encoder")
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@foss.st.com>
Acked-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Cc: <Stable@vger.kernel.org>
Link: https://lore.kernel.org/r/1614696235-24088-1-git-send-email-fabrice.gasnier@foss.st.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

When in SLAVE_MODE_DISABLED mode, the count still increases if the
counter is enabled because an internal clock is used. This patch fixes
the stm32_count_function_get() and stm32_count_function_set() functions
to properly handle this behavior.

Fixes: ad29937e206f ("counter: Add STM32 Timer quadrature encoder")
Cc: Fabrice Gasnier <fabrice.gasnier@st.com>
Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com>
Cc: Alexandre Torgue <alexandre.torgue@st.com>
Signed-off-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Reviewed-by: Fabrice Gasnier <fabrice.gasnier@foss.st.com>
Link: https://lore.kernel.org/r/20210226012931.161429-1-vilhelm.gray@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

The hardware doesn't support this. QPOSINIT is an initialization value
that is triggered by other things. When the counter overflows, it
always wraps around to zero.

Fixes: f213729f6796 "counter: new TI eQEP driver"
Signed-off-by: David Lechner <david@lechnology.com>
Acked-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Link: https://lore.kernel.org/r/20201214000927.1793062-1-david@lechnology.com
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

The ATMEL_TC_ETRGEDG_* defines are not masks but rather possible values
for CMR. This patch fixes the action_get() callback to properly check
for these values rather than mask them.

Fixes: 106b104137fd ("counter: Add microchip TCB capture counter")
Signed-off-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Acked-by: Kamel Bouhara <kamel.bouhara@bootlin.com>
Cc: <Stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20201114232805.253108-1-vilhelm.gray@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

The values given were the offset of the register after the last
register instead of the actual last register in each range. Fix
by using the correct last register of each range.

Fixes: f213729f6796 ("counter: new TI eQEP driver")
Signed-off-by: David Lechner <david@lechnology.com>
Acked-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Link: https://lore.kernel.org/r/20201025165122.607866-1-david@lechnology.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>