linux.git/drivers/staging/iio, branch v3.11

iio staging: fix lis3l02dq, read error handling

2013-07-09T21:11:53+00:00

Signed-off-by: Peter Meerwald 
Signed-off-by: Jonathan Cameron

staging:iio:ad7291: add missing .driver_module to struct iio_info

2013-07-06T09:39:26+00:00

Add missing .driver_module of struct iio_info. This prevents the
module from being removed from underneath its users.

Signed-off-by: Wei Yongjun 
Signed-off-by: Jonathan Cameron

iio: mxs-lradc: Remove useless check in read_raw

2013-07-06T09:34:51+00:00

The removed check in the read_raw implementation was always true,
therefore remove it. This also fixes a bug, by closely inspecting
the code, one can notice the iio_validate_scan_mask_onehot() will
always return 1 and therefore the subsequent condition will always
succeed, therefore making the mxs_lradc_read_raw() function always
return -EINVAL; .

Signed-off-by: Marek Vasut 
Tested-by: Otavio Salvador 
Acked-by: Hector Palacios 
Signed-off-by: Jonathan Cameron

iio: mxs-lradc: Fix misuse of iio->trig

2013-07-03T19:48:53+00:00

The struct iio_dev .trig field is to be used only by the IIO core,
the driver shall not fill this field. This fixes ugly crash when
the driver is compiled as a module and the module is rmmod'd.

Signed-off-by: Marek Vasut 
Cc: Fabio Estevam 
Cc: Jonathan Cameron 
Cc: Shawn Guo 
Signed-off-by: Jonathan Cameron

staging:iio:ad7291: Rework regulator handling

2013-06-14T20:55:17+00:00

Add platform_data for the driver which allows to specify whether an external
reference voltage is used or not. It is not possible to use the return value of
regulator_get() for this since the regulator framework is able to return a dummy
regulator in case no regulator has been specified. In this case the driver will
always get a valid regulator, no matter if it has been specified or not.  Also
make the regulator non-optional if platform_data states that an external
reference voltage should be used.

Signed-off-by: Lars-Peter Clausen 
Signed-off-by: Jonathan Cameron

staging:iio:ad7291: Use sign_extend32 instead of open-coding it

2013-06-14T20:54:09+00:00

This makes it a bit more obvious what is going on than open-coding it.

Signed-off-by: Lars-Peter Clausen 
Signed-off-by: Jonathan Cameron

staging:iio:ad7291: Use i2c_smbus_{read,write}_word_swapped instead of open-coding it

2013-06-14T20:53:31+00:00

No need to swap the upper and lower byte by hand if there is a helper function
which already does this.

Signed-off-by: Lars-Peter Clausen 
Signed-off-by: Jonathan Cameron

staging:iio:ad7291: Simplify threshold register lookup

2013-06-14T20:52:54+00:00

The AD7291 register map is nicely uniform and we can easily compute the
threshold register address for a certain channel using simple math instead of
using a look-up table.

Signed-off-by: Lars-Peter Clausen 
Signed-off-by: Jonathan Cameron

staging:iio:ad7291: Remove unnecessary dev_info() from probe()

2013-06-14T20:52:04+00:00

No need spam the log during probe if nothing went wrong.

Signed-off-by: Lars-Peter Clausen 
Signed-off-by: Jonathan Cameron

staging:iio:ad7291: Remove userspace reset

2013-06-14T20:51:28+00:00

There is no reason why userspace should want to trigger a manual reset of the
device, so remove this functionality.

Signed-off-by: Lars-Peter Clausen 
Signed-off-by: Jonathan Cameron