After a voltage brownout, sensor will now operate correctly.

Page 10 of ISL29018 data sheet and the Intersil Application Note 1534
describe the required initialization sequence:
1. Write 0x00 to register 0x08 (TEST)
2. Write 0x00 to register 0x00 (CMD1)
3. msleep(1)
4. program remaining registers as before
Signed-off-by: NGrant Grundler <grundler@chromium.org>
Reviewed-by: NBryan Freed <bfreed@chromium.org>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

The AD5933 is a high precision impedance converter system solution
that combines an on-board frequency generator with a 12-bit, 1 MSPS,
analog-to-digital converter (ADC). The frequency generator allows an
external complex impedance to be excited with a known frequency.

The response signal from the impedance is sampled by the on-board ADC
and a discrete Fourier transform (DFT) is processed by an on-chip DSP engine.
The DFT algorithm returns a real (R) and imaginary (I) data-word at each
output frequency.

Changes since V1:

Apply list review feedback:
Consistently use poll_time_jiffies.
Use be|le cpu endian helpers where applicable.
Add various comments.

Changes since V2:

Fix KernelVersion tag in Documentation.
Declare ad5933_default_pdata static.
Fix typos.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

The AD7280A monitoring system contains all the functions required for
general purpose monitoring and maintenance of stacked
lithium ion batteries as used in hybrid electric vehicles,
battery backup applications, etc.

Changes since V1:

Make cell channels all type IIO_IN_DIFF, update documentation accordingly.
Remove unused and redundant defines.
Use SI units where applicable.
Remove unnecessary wrapper function.
Remove redundant initialization.
Add comments where requested.
Revise event handler.
Use const where applicable.

Changes since V2:

Remove redundant adc.h include file, scheduled for removal.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Allow dynamic allocation and population of channel spec.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

The original definitions were duplicated to reduce tree churn during introduction of chan_spec
registration.  Now there is no point in maintaining the two sets of definitions.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Also get rid of a few function defs where they are only now in one core file
anyway.
Whilst here add mask = 0 to get rid of warning.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Another header unused after chan_spec conversions.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

The chan spec conversions and a small amount of pushdown of defines
mean this header is no longer used.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Not longer needed now all devices with gyro channels are using
iio_chan_spec registration.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Now all magnetomers are using iio_chan_spec there is no need for
the macros in magnet.h
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

iio_chan_spec conversion is straight forward.

Other changes:
* use i2c_smbus_write_byte_data in write_data
* c99 isms to setup various structures.
* move the comment about the scale factor and kill off trivial access func.
* make mode setting code use a bool rather than a ulong.

Ultimately I'd like to get rid of that mode bit entirely and handle it through
core pm routines, but have left it for now.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

I still have numerous issues with this driver and it
needs a substantial rewrite.

Whilst doing this patch at least one bug has been fixed (temp
reading didn't have an address) and a number of missing channel
info elements added.

Untested!
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Currently this is the only driver using some of the defines in accel.h.
If these become common we'll just add them to the channel_spec options,
for now push down into this driver.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Still a very odd driver. This patch has the side effect of changing
it's name slightly.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Simple conversion now we've dropped the type attributes.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This is a rare driver that is so simple it actually gets longer
as a result of this conversion.  Oh well, swings and roundabouts.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This needs confirmation that the devices is happy if another part is
talked to in between the request for a register and the read back.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Lots of minor bits and pieces.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

The equivalent should always be done using iio_alllocate_device
with to create a private area for the driver and then iio_priv
to access it.  There may be other uses for a private data pointer
but right now it just leads driver writers astray.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

There are some unusual corners in the probe function of this
driver, so may need another look.

V2: Now with the check for allocation success not inverted.
V3: Now with the i2c devdata calls actually being correctly cast.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Tested-by: NJon Brenner <jbrenner@TAOSinc.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Also push one down into the only driver that actually uses it.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Reorder to remove need for definitions currently in header.
Remove ida related utility function defs from header.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Fix trivial wrong index (no effect on any current drivers).
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Clear out all the excessive includes in the *_ring.c files.
Mixture of cut and paste bloat and things needed for code that
has gone away.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

I will assume these were cut and pastes messup.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

More headers now not needed after iio_chan_spec conversions.
A couple of drivers were using helpers from adc.h.

Given these were trivial, they are pushed down into the drivers.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Missing from the iio_chan_spec conversion patches.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

There is no clear use case for this functionality in a driver
that only supports slow reading via sysfs. The interface
use was non compliant with the abi motiving it being dropped.

It can go back in if anyone ever implements buffered reading
support for this device.  Then it will be controlled as
part of the buffering abi where this should indeed be correctly
supported.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

These should have gone with the iio_chan_spec conversion.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

These should have gone with the iio_chan_spec conversion patch.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Right now this results in increased code, but I still think it is worth doing to avoid
replication across instances of drivers etc and move as much stuff as possible to constant.
Ops structure is optional for the occasional driver that uses none of it (currently
only the ad7793).
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

These should really have gone away with the code that needed them.
Signed-off-by: NJonathan Cameron <jic23@cam.ac.uk>
Acked-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>