linux-toradex.git/drivers/iio, branch v4.4.24

include/linux/kernel.h: change abs() macro so it uses consistent return type

2016-09-30T08:18:33+00:00

commit 8f57e4d930d48217268315898212518d4d3e0773 upstream.

Rewrite abs() so that its return type does not depend on the
architecture and no unexpected type conversion happen inside of it.  The
only conversion is from unsigned to signed type.  char is left as a
return type but treated as a signed type regradless of it's actual
signedness.

With the old version, int arguments were promoted to long and depending
on architecture a long argument might result in s64 or long return type
(which may or may not be the same).

This came after some back and forth with Nicolas.  The current macro has
different return type (for the same input type) depending on
architecture which might be midly iritating.

An alternative version would promote to int like so:

	#define abs(x)	__abs_choose_expr(x, long long,			\
			__abs_choose_expr(x, long,			\
			__builtin_choose_expr(				\
				sizeof(x) <= sizeof(int),		\
				({ int __x = (x); __x<0?-__x:__x; }),	\
				((void)0))))

I have no preference but imagine Linus might.  :] Nicolas argument against
is that promoting to int causes iconsistent behaviour:

	int main(void) {
		unsigned short a = 0, b = 1, c = a - b;
		unsigned short d = abs(a - b);
		unsigned short e = abs(c);
		printf("%u %u\n", d, e);  // prints: 1 65535
	}

Then again, no sane person expects consistent behaviour from C integer
arithmetic.  ;)

Note:

  __builtin_types_compatible_p(unsigned char, char) is always false, and
  __builtin_types_compatible_p(signed char, char) is also always false.

Signed-off-by: Michal Nazarewicz 
Reviewed-by: Nicolas Pitre 
Cc: Srinivas Pandruvada 
Cc: Wey-Yi Guy 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Cc: Arnd Bergmann 
Signed-off-by: Greg Kroah-Hartman

iio: ensure ret is initialized to zero before entering do loop

2016-09-24T08:07:38+00:00

commit 5dba4b14bafe801083d01e1f400816df7e5a8f2e upstream.

A recent fix to iio_buffer_read_first_n_outer removed ret from being set by
a return from wait_event_interruptible and also added a continue in a loop
which causes the variable ret to not be set when it reaches the end of the
loop.  Fix this by initializing ret to zero.

Also remove extraneous white space at the end of the loop.

Fixes: fcf68f3c0bb2a5 ("fix sched WARNING "do not call blocking ops when !TASK_RUNNING")
Signed-off-by: Colin Ian King 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman

iio:core: fix IIO_VAL_FRACTIONAL sign handling

2016-09-24T08:07:38+00:00

commit 171c0091837c81ed5c949fec6966bb5afff2d1cf upstream.

7985e7c100 ("iio: Introduce a new fractional value type") introduced a
new IIO_VAL_FRACTIONAL value type meant to represent rational type numbers
expressed by a numerator and denominator combination.

Formating of IIO_VAL_FRACTIONAL values relies upon do_div() usage. This
fails handling negative values properly since parameters are reevaluated
as unsigned values.
Fix this by using div_s64_rem() instead. Computed integer part will carry
properly signed value. Formatted fractional part will always be positive.

Fixes: 7985e7c100 ("iio: Introduce a new fractional value type")
Signed-off-by: Gregor Boirie 
Reviewed-by: Lars-Peter Clausen 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman

iio: accel: kxsd9: Fix scaling bug

2016-09-24T08:07:38+00:00

commit 307fe9dd11ae44d4f8881ee449a7cbac36e1f5de upstream.

All the scaling of the KXSD9 involves multiplication with a
fraction number < 1.

However the scaling value returned from IIO_INFO_SCALE was
unpredictable as only the micros of the value was assigned, and
not the integer part, resulting in scaling like this:

$cat in_accel_scale
-1057462640.011978

Fix this by assigning zero to the integer part.

Tested-by: Jonathan Cameron 
Signed-off-by: Linus Walleij 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman

iio: fix pressure data output unit in hid-sensor-attributes

2016-09-24T08:07:38+00:00

commit 36afb176d3c9580651d7f410ed7f000ec48b5137 upstream.

According to IIO ABI definition, IIO_PRESSURE data output unit is
kilopascal:
http://lxr.free-electrons.com/source/Documentation/ABI/testing/sysfs-bus-iio

This patch fix output unit of HID pressure sensor IIO driver from pascal to
kilopascal to follow IIO ABI definition.

Signed-off-by: Kweh, Hock Leong 
Reviewed-by: Srinivas Pandruvada 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman

iio: accel: bmc150: reset chip at init time

2016-09-24T08:07:38+00:00

commit 1c500840934a138bd6b13556c210516e9301fbee upstream.

In at least one known setup, the chip comes up in a state where reading
the chip ID returns garbage unless it's been reset, due to noise on the
wires during system boot.

All supported chips have the same reset method, and based on the
datasheets they all need 1.3 or 1.8ms to recover after reset. So, do
the conservative thing here and always reset the chip.

Signed-off-by: Olof Johansson 
Reviewed-by: Srinivas Pandruvada 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman

iio: adc: at91: unbreak channel adc channel 3

2016-09-24T08:07:38+00:00

commit c2ab447454d498e709d9011c0f2d2945ee321f9b upstream.

The driver always assumes that an input device has been created when
reading channel 3. This causes a kernel panic when dereferencing
st->ts_input.

The change was introduced in
commit 84882b060301 ("iio: adc: at91_adc: Add support for touchscreens
without TSMR"). Earlier versions only entered that part of the if-else
statement if only the following flags are set:

AT91_ADC_IER_XRDY | AT91_ADC_IER_YRDY | AT91_ADC_IER_PRDY

Signed-off-by: Anders Darander 
Acked-by: Alexandre Belloni 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman

iio: ad799x: Fix buffered capture for ad7991/ad7995/ad7999

2016-09-24T08:07:38+00:00

commit 7d3cc21dab5313a02f2f3ca8164529b828a030d1 upstream.

The data buffer for captured mode for the ad799x driver is allocated in the
update_scan_mode() callback. This callback is not set in the iio_info
struct for the ad7791/ad7995/ad7999, which means that the data buffer is
not allocated when a captured transfer is started. As a result the driver
crashes when the first sample is received. To fix this properly set the
update_scan_mode() callback.

Fixes: d8dca33027c1 ("staging:iio:ad799x: Preallocate sample buffer")
Signed-off-by: Lars-Peter Clausen 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman

iio: adc: ti_am335x_adc: Increase timeout value waiting for ADC sample

2016-09-24T08:07:38+00:00

commit 7175cce1c3f1d8c8840d2004f78f96a3904249b5 upstream.

Now that open delay and sample delay for each channel is configurable
via DT, the default IDLE_TIMEOUT value is not enough as this is
calculated based on hardcoded macros. This results in driver returning
EBUSY sometimes. Fix this by increasing the timeout
value based on maximum value possible to open delay and sample delays
for each channel.

Fixes: 5dc11e810676e ("iio: adc: ti_am335x_adc: make sample delay, open delay, averaging DT parameters")
Signed-off-by: Vignesh R 
Acked-by: Lee Jones 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman

iio: adc: ti_am335x_adc: Protect FIFO1 from concurrent access

2016-09-24T08:07:38+00:00

commit 90c43ec6997a892448f1f86180a515f59cafd8a3 upstream.

It is possible that two or more ADC channels can be simultaneously
requested for raw samples, in which case there can be race in access to
FIFO data resulting in loss of samples.
If am335x_tsc_se_set_once() is called again from tiadc_read_raw(), when
ADC is still acquired to sample one of the channels, the second process
might be put into uninterruptible sleep state. Fix these issues, by
protecting FIFO access and channel configurations with a mutex. Since
tiadc_read_raw() might take anywhere between few microseconds to few
milliseconds to finish execution (depending on averaging and delay
values supplied via DT), its better to use mutex instead of spinlock.

Fixes: 7ca6740cd1cd4 ("mfd: input: iio: ti_amm335x: Rework TSC/ADC synchronization")
Signed-off-by: Vignesh R 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Greg Kroah-Hartman