linux-toradex.git/drivers/iio, branch v4.10-rc3

Merge tag 'iio-fixes-for-4.10a' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-linus

2017-01-02T15:59:44+00:00

Jonathan writes:

First round of IIO fixes for the 4.10 cycle.

* 104-quad-8
  - Fix selecting wrong register when the index control register is desired.
  - Fix an off by one error when addressing the input/output control register.
  - Fix inverted logic on the active high / low control
* bmi160
  - Sleep for worst case rather than best case amount of time after cmd
  execution begins.
* max44000
  - typo fix in illuminance_integration_time_available listing.
* st-sensors
  - Fix channel data passing.  This one took a while to get tested on 24bit
  parts. Definitely one for stable asap as the bug broke quite a few parts.
  - lis3lv02 needs a data alignment bit set and the scaling was wrong.
* ti_am335x
  - depend on HAS_DMA

iio: accel: st_accel: fix LIS3LV02 reading and scaling

2016-12-31T14:35:04+00:00

The LIS3LV02 has a special bit that need to be set to get the
read values left aligned. Before this patch we get gibberish
like this:

iio_generic_buffer -a -c10 -n lis3lv02dl_accel
(...)
0.000000 -0.010042 -0.642688 19155832931907
0.000000 -0.010042 -0.642688 19155858751073

Which is because we read a raw value for 1g as 64 which is
the nominal 1024 for 1g shifted 4 bits to the left by being
right-aligned rather than left aligned.

Since all other sensors are left aligned, add some code to
set the special DAS (data alignment setting) bit to 1 so that
the right value is now read like this:

iio_generic_buffer -a -c10 -n lis3lv02dl_accel
(...)
0.000000 -0.147095 -10.120135 24761614364956
-0.029419 -0.176514 -10.120135 24761631624540

The scaling was weird as well: we have a gain of 1000 for 1g
and 3000 for 6g. I don't even remember how I came up with the
old values but they are wrong.

Fixes: 3acddf74f807 ("iio: st-sensors: add support for lis3lv02d accelerometer")
Cc: Lorenzo Bianconi 
Cc: Giuseppe Barba 
Cc: Denis Ciocca 
Signed-off-by: Linus Walleij 
Signed-off-by: Jonathan Cameron

iio: common: st_sensors: fix channel data parsing

2016-12-31T14:32:09+00:00

Using realbits as i2c/spi read len, when that value is not byte aligned
(e.g 12 bits), lead to skip msb part of out data registers.
Fix this taking into account scan_type.shift in addition to
scan_type.realbits as read length:

read_len = DIV_ROUND_UP(realbits + shift, 8)

This fix has been tested on 8, 12, 16, 24 bit sensors

Fixes: e7385de5291e ("iio:st_sensors: align on storagebits boundaries")
Signed-off-by: Lorenzo Bianconi 
Tested-by: Linus Walleij 
Cc: 
Signed-off-by: Jonathan Cameron

iio: max44000: correct value in illuminance_integration_time_available

2016-12-30T20:20:23+00:00

According to the datasheet, the shortest available integration time for
ALS ADC conversion is 1.5625ms but illuminance_integration_time_available
sysfs file shows wrong value.

Cc: Crestez Dan Leonard 
Cc: Jonathan Cameron 
Cc: Hartmut Knaack 
Cc: Lars-Peter Clausen 
Cc: Peter Meerwald-Stadler 
Signed-off-by: Akinobu Mita 
Fixes: d5d8f49b6 ("max44000: Expose ambient sensor scaling")
Cc: 
Signed-off-by: Jonathan Cameron

iio: adc: TI_AM335X_ADC should depend on HAS_DMA

2016-12-30T16:37:48+00:00

If NO_DMA=y:

    ERROR: "bad_dma_ops" [drivers/iio/adc/ti_am335x_adc.ko] undefined!

Add a dependency on HAS_DMA to fix this.

Signed-off-by: Geert Uytterhoeven 
Fixes: f438b9da (" drivers: iio: ti_am335x_adc: add dma support")
Cc: 
Signed-off-by: Jonathan Cameron

iio: bmi160: Fix time needed to sleep after command execution

2016-12-30T10:52:00+00:00

Datasheet specifies typical and maximum execution times for which CMD
register is occupied after previous command execution. We took these
values as minimum and maximum time for usleep_range() call before making
a new command execution.

To be sure, that the CMD register is no longer occupied we need to wait
*at least* the maximum time specified by datasheet.

Signed-off-by: Marcin Niestroj 
Cc: 
Signed-off-by: Jonathan Cameron

ktime: Cleanup ktime_set() usage

2016-12-25T16:21:22+00:00

ktime_set(S,N) was required for the timespec storage type and is still
useful for situations where a Seconds and Nanoseconds part of a time value
needs to be converted. For anything where the Seconds argument is 0, this
is pointless and can be replaced with a simple assignment.

Signed-off-by: Thomas Gleixner 
Cc: Peter Zijlstra

iio: 104-quad-8: Fix active level mismatch for the preset enable option

2016-12-03T10:08:09+00:00

The set_to_preset_on_index sysfs attribute provides a boolean
configuration option to enable a preset operation on the respective
channel's counter when Index occurs. However, the corresponding
configuration bit on the Input/Output Control register must be set low
to enable the preset operation.

This patch inverts the internal driver boolean logic exposed via the
set_to_preset_on_index attribute in order to correctly configure the
device which expects an active low bit.

Fixes: 28e5d3bb0325 ("iio: 104-quad-8: Add IIO support for the ACCES 104-QUAD-8")
Signed-off-by: William Breathitt Gray 
Cc: 
Signed-off-by: Jonathan Cameron

iio: 104-quad-8: Fix off-by-one errors when addressing IOR

2016-12-03T10:03:07+00:00

The Input/Output Control register (IOR) is offset by 1 from the
respective channel data register. This patch fixes off-by-one errors
when attempting to write to a channel IOR where the base address was not
properly offset.

Fixes: 28e5d3bb0325 ("iio: 104-quad-8: Add IIO support for the ACCES 104-QUAD-8")
Signed-off-by: William Breathitt Gray 
Cc: 
Signed-off-by: Jonathan Cameron

iio: 104-quad-8: Fix index control configuration

2016-12-03T10:00:24+00:00

The LS7266R1 requires bits 5 & 6 to be high in order to select the Index
Control Register. This patch fixes a typo that incorrectly selects the
Input/Output Control Register where the Index Control Register was
desired.

Fixes: 28e5d3bb0325 ("iio: 104-quad-8: Add IIO support for the ACCES 104-QUAD-8")
Signed-off-by: William Breathitt Gray 
Cc: 
Signed-off-by: Jonathan Cameron