linux-toradex.git/drivers/iio, branch v3.18.16

iio: adc: xilinx: Fix VREFN sign

2015-06-09T18:31:41+00:00

[ Upstream commit 97ffae1d30c3f6ceee67d5b0d3e540c08c13c744 ]

The VREFN channel is bipolar, not unipolar. Small negative values do
occur (e.g., -1mV), and unsigned conversion maps them incorrectly to
large positive values (about +1V), so fix this.

Signed-off-by: Thomas Betker 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: adc: xilinx: Fix VREFP scale

2015-06-09T18:31:41+00:00

[ Upstream commit 00db4e52f4541965f7fda225eb458a75f892017b ]

The scaling factor for VREFP is 3.0/4096, not 1.0/4096; fix this to get
correct readings.

Signed-off-by: Thomas Betker 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: adc: xilinx: Fix "vccaux" channel .address

2015-06-09T18:31:40+00:00

[ Upstream commit d6c96c42283601e311a7a1a3d7e51cde9d7fdb6e ]

For the "vccaux" channel, read the VCCAUX register, not VCCINT.

Signed-off-by: Thomas Betker 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: adc: xilinx: Fix register addresses

2015-06-09T18:31:39+00:00

[ Upstream commit 3960d2c0c4aafe98da47a4a2eb64dfa8e88d8df5 ]

Define the register addresses for MIN_VCCPINT, MIN_VCCPAUX, MIN_VCCO_DDR
correctly.

Signed-off-by: Thomas Betker 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: pressure: hid-sensor-press: Fix modifier

2015-06-09T18:31:39+00:00

[ Upstream commit 964e2255f1d73fc0136bc206a78a1f86bdad72a7 ]

Fix "null" in the raw attribute and scan elements.

Signed-off-by: Srinivas Pandruvada 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: light: hid-sensor-prox: Fix modifier

2015-06-09T18:31:38+00:00

[ Upstream commit c2aab3d58b96002555a3e70004f593b043830248 ]

Currently in_proximity_(null)_raw is getting presented as raw sysfs
attribute. Same with the scan_elements.
The modifier doesn't apply to this channel.

Signed-off-by: Srinivas Pandruvada 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: imu: Use iio_trigger_get for indio_dev->trig assignment

2015-04-24T21:13:53+00:00

[ Upstream commit 4ce7ca89d6e8eae9e201cd0e972ba323f33e2fb4 ]

This patch uses iio_trigger_get to increment the reference
count of trigger device, to avoid incorrect assignment.
Can result in a null pointer dereference during removal if the
trigger has been changed before removal.

This patch refers to a similar situation encountered through the
following discussion:
http://www.spinics.net/lists/linux-iio/msg13669.html

Signed-off-by: Darshana Padmadas 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: adc: vf610: use ADC clock within specification

2015-04-24T21:13:52+00:00

[ Upstream commit f54e9f2be312a4e71b54aea865b2e33ccb95ef0c ]

Depending on conversion mode used, the ADC clock (ADCK) needs
to be below a maximum frequency. According to Vybrid's data
sheet this is 20MHz for the low power conversion mode.

The ADC clock is depending on input clock, which is the bus
clock by default. Vybrid SoC are typically clocked at at 400MHz
or 500MHz, which leads to 66MHz or 83MHz bus clock respectively.
Hence, a divider of 8 is required to stay below the specified
maximum clock of 20MHz.

Due to the different bus clock speeds, the resulting sampling
frequency is not static. Hence use the ADC clock and calculate
the actual available sampling frequency dynamically.

This fixes bogous values observed on some 500MHz clocked Vybrid
SoC. The resulting value usually showed Bit 9 being stuck at 1,
or 0, which lead to a value of +/-512.

Signed-off-by: Stefan Agner 
Acked-by: Fugang Duan 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: bmc150: change sampling frequency

2015-04-24T21:13:51+00:00

[ Upstream commit 0ba8da961bd868c67a8dae3dbbee145514515e9c ]

Currently driver reports device bandwidth list as available
sampling frequency. But sampling frequency is actually twice
the device bandwidth. This patch fixes this issue.

Signed-off-by: Sathyanarayanan Kuppuswamy 
Signed-off-by: Octavian Purdila 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin

iio: core: Fix double free.

2015-04-24T21:13:50+00:00

[ Upstream commit c1b03ab5e886760bdd38c9c7a27af149046ffe01 ]

When an error occurred during event registration memory was freed twice
resulting in kernel memory corruption and a crash in unrelated code.

The problem was caused by
	iio_device_unregister_eventset()
	iio_device_unregister_sysfs()

being called twice, once on the error path and then
again via iio_dev_release().

Fix this by making these two functions idempotent so they
may be called multiple times.

The problem was observed before applying
	78b33216 iio:core: Handle error when mask type is not separate

Signed-off-by: Martin Fuzzey 
Cc: 
Signed-off-by: Jonathan Cameron 
Signed-off-by: Sasha Levin