mbox series

[PULL,v3] Second set of new device support, cleanups etc for IIO in the 5.4 cycle.

Message ID 20190903211708.2ba07358@archlinux (mailing list archive)
State New, archived
Headers show
Series [PULL,v3] Second set of new device support, cleanups etc for IIO in the 5.4 cycle. | expand

Pull-request

git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git tags/iio-for-5.4b-take3

Message

Jonathan Cameron Sept. 3, 2019, 8:17 p.m. UTC
The following changes since commit 81e0da703fdba1ee126868bf8350592c79cdba13:

  staging: sm750fb: fix odd license text (2019-08-15 14:43:57 +0200)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git tags/iio-for-5.4b-take3

for you to fetch changes up to 6fa029486b74c266c39b718b8643ed42d24f977a:

  iio: imu: st_lsm6dsx: rely on IIO_G_TO_M_S_2 for gain definition for LSM9DS1 (2019-09-03 21:11:28 +0100)

----------------------------------------------------------------
Second set of new device support, cleanups and features for IIO in the 5.4 cycle

Revised pull request to fix up a missing Signed-off-by and roll in
a fix in the lsm9ds1 support after I broke it when applying.
Revised again because the fix changed a hash meaning a fix
that previously followed it now had the wrong fixes tag.

A few fixes in here that could have gone a faster path but aren't quite
worth the rush for 5.3.

New device support
* ad7606
  - Support the ad7606b which adds a software controlled mode alongside
    the pin controlled only approach of the ad7606. Including dt-bindings.
* lsm6dsx
  - Add support for the gyro and accelerometer part of the lsm9ds1 which is
    a compound device also including a magnetometer (st_sensors driver).
    Includes bindings and precursor rework of the driver.

Features
* ad7192
  - Add support for low pass filter control.
  - DT binding docs.

Cleanups and minor fixes
* MAINTAINERS
  - Fix a typo in a path.
  - Add entry for ad7606
* ad5380
  - Fix a failure to dereference a pointer before atempting to assign the
    value.
* ad7192
  - Drop platform data as not used in mainline and we now have full DT bindings.
* ad7606
  - YAML conversion for dt-bindings.
* adis16240
  - Rework write_raw to make it more readable using GENMASK.
* adis16460
  - Fix and issue with an unsigned variable holding potential negatives.
* cros_ec
  - Fix missing default of calibration vector so that we get 'something'
    before calibration is complete on a given axis.
* hid-sensors
  - Use int_pow instead of opencoding.
* isl29501
  - rename dt-binding docs to include renesas inline with other renesas parts
    and general current convention.
* kxcjk1013
  - Improve comments on the 'unusual' ACPI ids used to identify which sensor
    is which in certain laptops.
* lsm6dsx
  - Add one bit to the fifo status masks for a number of parts.
  - Drop a reserved entry from the sensitivity values to tidy up interface.
  - Use core conversion macro from G to m/s^2 for lsm9ds1 to make it easier
    to relate to the datasheet and consistent with other parts supported.
* max1027
  - Use device managed APIs to avoid manual error handling and cleanup.
* rfd77402
  - Typo in Kconfig help.
* sc27xx
  - Switch to polling mode from interrupts as interrupt handling typically
    to slow for very short sleeps.
* st-sensors
  - Fix some missing selects for regmap.
* tools
  - Add a .gitignore containing the binary outputs.

----------------------------------------------------------------
Alexandru Ardelean (1):
      iio: imu: adis16460: fix variable signedness

Andy Shevchenko (1):
      iio: hid-sensor-attributes: Convert to use int_pow()

Bartosz Golaszewski (1):
      tools: iio: add .gitignore

Beniamin Bia (3):
      MAINTAINERS: Add Beniamin Bia for AD7606 driver
      dt-bindings: iio: adc: Migrate AD7606 documentation to yaml
      dt-bindings: iio: adc: Add AD7606B ADC documentation

Christophe JAILLET (1):
      Kconfig: Fix the reference to the RFD77402 ToF sensor in the 'help' section

Chuhong Yuan (1):
      iio: adc: max1027: Use device-managed APIs

Colin Ian King (1):
      iio: dac: ad5380: fix incorrect assignment to val

Denis Efremov (1):
      MAINTAINERS: ftm-quaddec: Fix typo in a filepath

Freeman Liu (1):
      iio: adc: sc27xx: Change to polling mode to read data

Gwendal Grignou (1):
      iio: cros_ec: set calibscale for 3d MEMS to unit vector

Hans de Goede (1):
      iio: accel: kxcjk1013: Improve comments on KIOX010A and KIOX020A ACPI ids

Lorenzo Bianconi (3):
      iio: imu: st_lsm6dsx: introduce update_fifo function pointer
      iio: imu: st_lsm6dsx: remove invalid gain value for LSM9DS1
      iio: imu: st_lsm6dsx: rely on IIO_G_TO_M_S_2 for gain definition for LSM9DS1

Martin Kepplinger (3):
      iio: imu: st_lsm6dsx: move register definitions to sensor_settings struct
      iio: imu: st_lsm6dsx: add support for accel/gyro unit of lsm9ds1
      dt-bindings: iio: imu: st_lsm6dsx: add lsm9ds1 device bindings

Mircea Caprioru (3):
      staging: iio: adc: ad7192: Remove platform data
      staging: iio: adc: ad7192: Add low_pass_3db_filter_frequency
      dt-bindings: iio: adc: ad7192: Add binding documentation for AD7192

Rodrigo Ribeiro (1):
      staging: iio: accel: adis16240: Improve readability on write_raw function

Simon Horman (1):
      dt-bindings: iio: light: isl29501: Rename bindings documentation file

Stefan Popa (1):
      iio: adc: ad7606: Add support for AD7606B ADC

YueHaibing (1):
      iio: st_sensors: Fix build error

mario tesi (1):
      iio: imu: st_lsm6dsx: Fix FIFO diff mask for tagged fifo

 .../devicetree/bindings/iio/adc/adi,ad7192.yaml    | 121 +++++++++++++++
 .../devicetree/bindings/iio/adc/adi,ad7606.txt     |  66 ---------
 .../devicetree/bindings/iio/adc/adi,ad7606.yaml    | 138 +++++++++++++++++
 .../devicetree/bindings/iio/imu/st_lsm6dsx.txt     |   1 +
 .../light/{isl29501.txt => renesas,isl29501.txt}   |   0
 MAINTAINERS                                        |   5 +-
 drivers/iio/accel/kxcjk-1013.c                     |   4 +-
 drivers/iio/adc/ad7606.c                           |  13 +-
 drivers/iio/adc/ad7606.h                           |   4 +
 drivers/iio/adc/ad7606_spi.c                       | 109 +++++++++++++-
 drivers/iio/adc/max1027.c                          |  38 +----
 drivers/iio/adc/sc27xx_adc.c                       |  81 ++++------
 .../common/cros_ec_sensors/cros_ec_sensors_core.c  |   5 +-
 .../iio/common/hid-sensors/hid-sensor-attributes.c |  53 +++----
 drivers/iio/common/st_sensors/Kconfig              |   2 +
 drivers/iio/dac/ad5380.c                           |   2 +-
 drivers/iio/imu/adis16460.c                        |   2 +-
 drivers/iio/imu/st_lsm6dsx/Kconfig                 |   2 +-
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx.h            |  12 ++
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c     |  19 ++-
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c       | 139 ++++++++++++++++--
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_i2c.c        |   5 +
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_spi.c        |   5 +
 drivers/iio/proximity/Kconfig                      |   2 +-
 drivers/staging/iio/accel/adis16240.c              |   5 +-
 drivers/staging/iio/adc/ad7192.c                   | 163 +++++++++++++++++----
 drivers/staging/iio/adc/ad7192.h                   |  37 -----
 tools/iio/.gitignore                               |   4 +
 28 files changed, 756 insertions(+), 281 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/iio/adc/adi,ad7192.yaml
 delete mode 100644 Documentation/devicetree/bindings/iio/adc/adi,ad7606.txt
 create mode 100644 Documentation/devicetree/bindings/iio/adc/adi,ad7606.yaml
 rename Documentation/devicetree/bindings/iio/light/{isl29501.txt => renesas,isl29501.txt} (100%)
 delete mode 100644 drivers/staging/iio/adc/ad7192.h
 create mode 100644 tools/iio/.gitignore

Comments

Greg KH Sept. 3, 2019, 8:28 p.m. UTC | #1
On Tue, Sep 03, 2019 at 09:17:29PM +0100, Jonathan Cameron wrote:
> The following changes since commit 81e0da703fdba1ee126868bf8350592c79cdba13:
> 
>   staging: sm750fb: fix odd license text (2019-08-15 14:43:57 +0200)
> 
> are available in the Git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git tags/iio-for-5.4b-take3

This one worked :)

Now pulled and pushed out, thanks.

greg k-h
Nathan Chancellor Sept. 5, 2019, 2:50 a.m. UTC | #2
On Tue, Sep 03, 2019 at 09:17:29PM +0100, Jonathan Cameron wrote:
> The following changes since commit 81e0da703fdba1ee126868bf8350592c79cdba13:
> 
>   staging: sm750fb: fix odd license text (2019-08-15 14:43:57 +0200)
> 
> are available in the Git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git tags/iio-for-5.4b-take3
> 
> for you to fetch changes up to 6fa029486b74c266c39b718b8643ed42d24f977a:
> 
>   iio: imu: st_lsm6dsx: rely on IIO_G_TO_M_S_2 for gain definition for LSM9DS1 (2019-09-03 21:11:28 +0100)
> 
> ----------------------------------------------------------------
> Second set of new device support, cleanups and features for IIO in the 5.4 cycle
> 
> Revised pull request to fix up a missing Signed-off-by and roll in
> a fix in the lsm9ds1 support after I broke it when applying.
> Revised again because the fix changed a hash meaning a fix
> that previously followed it now had the wrong fixes tag.
> 
> A few fixes in here that could have gone a faster path but aren't quite
> worth the rush for 5.3.
> 
> New device support
> * ad7606
>   - Support the ad7606b which adds a software controlled mode alongside
>     the pin controlled only approach of the ad7606. Including dt-bindings.
> * lsm6dsx
>   - Add support for the gyro and accelerometer part of the lsm9ds1 which is
>     a compound device also including a magnetometer (st_sensors driver).
>     Includes bindings and precursor rework of the driver.
> 
> Features
> * ad7192
>   - Add support for low pass filter control.
>   - DT binding docs.
> 
> Cleanups and minor fixes
> * MAINTAINERS
>   - Fix a typo in a path.
>   - Add entry for ad7606
> * ad5380
>   - Fix a failure to dereference a pointer before atempting to assign the
>     value.
> * ad7192
>   - Drop platform data as not used in mainline and we now have full DT bindings.
> * ad7606
>   - YAML conversion for dt-bindings.
> * adis16240
>   - Rework write_raw to make it more readable using GENMASK.
> * adis16460
>   - Fix and issue with an unsigned variable holding potential negatives.
> * cros_ec
>   - Fix missing default of calibration vector so that we get 'something'
>     before calibration is complete on a given axis.
> * hid-sensors
>   - Use int_pow instead of opencoding.
> * isl29501
>   - rename dt-binding docs to include renesas inline with other renesas parts
>     and general current convention.
> * kxcjk1013
>   - Improve comments on the 'unusual' ACPI ids used to identify which sensor
>     is which in certain laptops.
> * lsm6dsx
>   - Add one bit to the fifo status masks for a number of parts.
>   - Drop a reserved entry from the sensitivity values to tidy up interface.
>   - Use core conversion macro from G to m/s^2 for lsm9ds1 to make it easier
>     to relate to the datasheet and consistent with other parts supported.
> * max1027
>   - Use device managed APIs to avoid manual error handling and cleanup.
> * rfd77402
>   - Typo in Kconfig help.
> * sc27xx
>   - Switch to polling mode from interrupts as interrupt handling typically
>     to slow for very short sleeps.
> * st-sensors
>   - Fix some missing selects for regmap.
> * tools
>   - Add a .gitignore containing the binary outputs.
> 
> ----------------------------------------------------------------
> Alexandru Ardelean (1):
>       iio: imu: adis16460: fix variable signedness
> 
> Andy Shevchenko (1):
>       iio: hid-sensor-attributes: Convert to use int_pow()

This patch causes a linking failure on arm32 when CONFIG_HID_SENSOR_HUB
and CONFIG_HID_SENSOR_IIO_COMMON are set (I caught this on arm32
allyesconfig):

arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o: in function `hid_sensor_read_raw_hyst_value':
hid-sensor-attributes.c:(.text+0x23c): undefined reference to `__aeabi_uldivmod'
arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x280): undefined reference to `__aeabi_uldivmod'
arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o: in function `hid_sensor_format_scale':
hid-sensor-attributes.c:(.text+0x388): undefined reference to `__aeabi_uldivmod'
arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x3b0): undefined reference to `__aeabi_uldivmod'
arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x414): undefined reference to `__aeabi_uldivmod'
arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o:hid-sensor-attributes.c:(.text+0x444): more undefined references to `__aeabi_uldivmod' follow

int_pow() returns u64, which causes issues when just trying to use
normal division. Looks like most of the dividends are 32-bit so not
really sure how to resolve this otherwise I would have sent a patch.
Please look into this when you get a chance.

Cheers,
Nathan
Greg KH Sept. 5, 2019, 5:34 a.m. UTC | #3
On Wed, Sep 04, 2019 at 07:50:08PM -0700, Nathan Chancellor wrote:
> On Tue, Sep 03, 2019 at 09:17:29PM +0100, Jonathan Cameron wrote:
> > The following changes since commit 81e0da703fdba1ee126868bf8350592c79cdba13:
> > 
> >   staging: sm750fb: fix odd license text (2019-08-15 14:43:57 +0200)
> > 
> > are available in the Git repository at:
> > 
> >   git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git tags/iio-for-5.4b-take3
> > 
> > for you to fetch changes up to 6fa029486b74c266c39b718b8643ed42d24f977a:
> > 
> >   iio: imu: st_lsm6dsx: rely on IIO_G_TO_M_S_2 for gain definition for LSM9DS1 (2019-09-03 21:11:28 +0100)
> > 
> > ----------------------------------------------------------------
> > Second set of new device support, cleanups and features for IIO in the 5.4 cycle
> > 
> > Revised pull request to fix up a missing Signed-off-by and roll in
> > a fix in the lsm9ds1 support after I broke it when applying.
> > Revised again because the fix changed a hash meaning a fix
> > that previously followed it now had the wrong fixes tag.
> > 
> > A few fixes in here that could have gone a faster path but aren't quite
> > worth the rush for 5.3.
> > 
> > New device support
> > * ad7606
> >   - Support the ad7606b which adds a software controlled mode alongside
> >     the pin controlled only approach of the ad7606. Including dt-bindings.
> > * lsm6dsx
> >   - Add support for the gyro and accelerometer part of the lsm9ds1 which is
> >     a compound device also including a magnetometer (st_sensors driver).
> >     Includes bindings and precursor rework of the driver.
> > 
> > Features
> > * ad7192
> >   - Add support for low pass filter control.
> >   - DT binding docs.
> > 
> > Cleanups and minor fixes
> > * MAINTAINERS
> >   - Fix a typo in a path.
> >   - Add entry for ad7606
> > * ad5380
> >   - Fix a failure to dereference a pointer before atempting to assign the
> >     value.
> > * ad7192
> >   - Drop platform data as not used in mainline and we now have full DT bindings.
> > * ad7606
> >   - YAML conversion for dt-bindings.
> > * adis16240
> >   - Rework write_raw to make it more readable using GENMASK.
> > * adis16460
> >   - Fix and issue with an unsigned variable holding potential negatives.
> > * cros_ec
> >   - Fix missing default of calibration vector so that we get 'something'
> >     before calibration is complete on a given axis.
> > * hid-sensors
> >   - Use int_pow instead of opencoding.
> > * isl29501
> >   - rename dt-binding docs to include renesas inline with other renesas parts
> >     and general current convention.
> > * kxcjk1013
> >   - Improve comments on the 'unusual' ACPI ids used to identify which sensor
> >     is which in certain laptops.
> > * lsm6dsx
> >   - Add one bit to the fifo status masks for a number of parts.
> >   - Drop a reserved entry from the sensitivity values to tidy up interface.
> >   - Use core conversion macro from G to m/s^2 for lsm9ds1 to make it easier
> >     to relate to the datasheet and consistent with other parts supported.
> > * max1027
> >   - Use device managed APIs to avoid manual error handling and cleanup.
> > * rfd77402
> >   - Typo in Kconfig help.
> > * sc27xx
> >   - Switch to polling mode from interrupts as interrupt handling typically
> >     to slow for very short sleeps.
> > * st-sensors
> >   - Fix some missing selects for regmap.
> > * tools
> >   - Add a .gitignore containing the binary outputs.
> > 
> > ----------------------------------------------------------------
> > Alexandru Ardelean (1):
> >       iio: imu: adis16460: fix variable signedness
> > 
> > Andy Shevchenko (1):
> >       iio: hid-sensor-attributes: Convert to use int_pow()
> 
> This patch causes a linking failure on arm32 when CONFIG_HID_SENSOR_HUB
> and CONFIG_HID_SENSOR_IIO_COMMON are set (I caught this on arm32
> allyesconfig):
> 
> arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o: in function `hid_sensor_read_raw_hyst_value':
> hid-sensor-attributes.c:(.text+0x23c): undefined reference to `__aeabi_uldivmod'
> arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x280): undefined reference to `__aeabi_uldivmod'
> arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o: in function `hid_sensor_format_scale':
> hid-sensor-attributes.c:(.text+0x388): undefined reference to `__aeabi_uldivmod'
> arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x3b0): undefined reference to `__aeabi_uldivmod'
> arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x414): undefined reference to `__aeabi_uldivmod'
> arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o:hid-sensor-attributes.c:(.text+0x444): more undefined references to `__aeabi_uldivmod' follow
> 
> int_pow() returns u64, which causes issues when just trying to use
> normal division. Looks like most of the dividends are 32-bit so not
> really sure how to resolve this otherwise I would have sent a patch.
> Please look into this when you get a chance.

0-day is now reporting this as well :(
Jonathan Cameron Sept. 5, 2019, 9:11 a.m. UTC | #4
On Thu, 5 Sep 2019 07:34:49 +0200
Greg KH <gregkh@linuxfoundation.org> wrote:

> On Wed, Sep 04, 2019 at 07:50:08PM -0700, Nathan Chancellor wrote:
> > On Tue, Sep 03, 2019 at 09:17:29PM +0100, Jonathan Cameron wrote:  
> > > The following changes since commit 81e0da703fdba1ee126868bf8350592c79cdba13:
> > > 
> > >   staging: sm750fb: fix odd license text (2019-08-15 14:43:57 +0200)
> > > 
> > > are available in the Git repository at:
> > > 
> > >   git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git tags/iio-for-5.4b-take3
> > > 
> > > for you to fetch changes up to 6fa029486b74c266c39b718b8643ed42d24f977a:
> > > 
> > >   iio: imu: st_lsm6dsx: rely on IIO_G_TO_M_S_2 for gain definition for LSM9DS1 (2019-09-03 21:11:28 +0100)
> > > 
> > > ----------------------------------------------------------------
> > > Second set of new device support, cleanups and features for IIO in the 5.4 cycle
> > > 
> > > Revised pull request to fix up a missing Signed-off-by and roll in
> > > a fix in the lsm9ds1 support after I broke it when applying.
> > > Revised again because the fix changed a hash meaning a fix
> > > that previously followed it now had the wrong fixes tag.
> > > 
> > > A few fixes in here that could have gone a faster path but aren't quite
> > > worth the rush for 5.3.
> > > 
> > > New device support
> > > * ad7606
> > >   - Support the ad7606b which adds a software controlled mode alongside
> > >     the pin controlled only approach of the ad7606. Including dt-bindings.
> > > * lsm6dsx
> > >   - Add support for the gyro and accelerometer part of the lsm9ds1 which is
> > >     a compound device also including a magnetometer (st_sensors driver).
> > >     Includes bindings and precursor rework of the driver.
> > > 
> > > Features
> > > * ad7192
> > >   - Add support for low pass filter control.
> > >   - DT binding docs.
> > > 
> > > Cleanups and minor fixes
> > > * MAINTAINERS
> > >   - Fix a typo in a path.
> > >   - Add entry for ad7606
> > > * ad5380
> > >   - Fix a failure to dereference a pointer before atempting to assign the
> > >     value.
> > > * ad7192
> > >   - Drop platform data as not used in mainline and we now have full DT bindings.
> > > * ad7606
> > >   - YAML conversion for dt-bindings.
> > > * adis16240
> > >   - Rework write_raw to make it more readable using GENMASK.
> > > * adis16460
> > >   - Fix and issue with an unsigned variable holding potential negatives.
> > > * cros_ec
> > >   - Fix missing default of calibration vector so that we get 'something'
> > >     before calibration is complete on a given axis.
> > > * hid-sensors
> > >   - Use int_pow instead of opencoding.
> > > * isl29501
> > >   - rename dt-binding docs to include renesas inline with other renesas parts
> > >     and general current convention.
> > > * kxcjk1013
> > >   - Improve comments on the 'unusual' ACPI ids used to identify which sensor
> > >     is which in certain laptops.
> > > * lsm6dsx
> > >   - Add one bit to the fifo status masks for a number of parts.
> > >   - Drop a reserved entry from the sensitivity values to tidy up interface.
> > >   - Use core conversion macro from G to m/s^2 for lsm9ds1 to make it easier
> > >     to relate to the datasheet and consistent with other parts supported.
> > > * max1027
> > >   - Use device managed APIs to avoid manual error handling and cleanup.
> > > * rfd77402
> > >   - Typo in Kconfig help.
> > > * sc27xx
> > >   - Switch to polling mode from interrupts as interrupt handling typically
> > >     to slow for very short sleeps.
> > > * st-sensors
> > >   - Fix some missing selects for regmap.
> > > * tools
> > >   - Add a .gitignore containing the binary outputs.
> > > 
> > > ----------------------------------------------------------------
> > > Alexandru Ardelean (1):
> > >       iio: imu: adis16460: fix variable signedness
> > > 
> > > Andy Shevchenko (1):
> > >       iio: hid-sensor-attributes: Convert to use int_pow()  
> > 
> > This patch causes a linking failure on arm32 when CONFIG_HID_SENSOR_HUB
> > and CONFIG_HID_SENSOR_IIO_COMMON are set (I caught this on arm32
> > allyesconfig):
> > 
> > arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o: in function `hid_sensor_read_raw_hyst_value':
> > hid-sensor-attributes.c:(.text+0x23c): undefined reference to `__aeabi_uldivmod'
> > arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x280): undefined reference to `__aeabi_uldivmod'
> > arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o: in function `hid_sensor_format_scale':
> > hid-sensor-attributes.c:(.text+0x388): undefined reference to `__aeabi_uldivmod'
> > arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x3b0): undefined reference to `__aeabi_uldivmod'
> > arm-linux-gnueabi-ld: hid-sensor-attributes.c:(.text+0x414): undefined reference to `__aeabi_uldivmod'
> > arm-linux-gnueabi-ld: drivers/iio/common/hid-sensors/hid-sensor-attributes.o:hid-sensor-attributes.c:(.text+0x444): more undefined references to `__aeabi_uldivmod' follow
> > 
> > int_pow() returns u64, which causes issues when just trying to use
> > normal division. Looks like most of the dividends are 32-bit so not
> > really sure how to resolve this otherwise I would have sent a patch.
> > Please look into this when you get a chance.  
> 
> 0-day is now reporting this as well :(

Yes sorry about this.  I'd failed to chase up on why 0-day isn't building
iio.git/testing any more which I'd come to rely on for 32bit coverage.
I have a question out now on that + whether we should just revert this patch
and revisit in the next cycle.  

Thanks,

Jonathan