staging: iio: tsl2x7x: use auto increment I2C protocol

The hardware supports 16-bit ALS and proximity readings, however the
datasheet recommends using the I2C auto increment protocol so that the
correct high and low bytes are read even if the integration cycle ends
between reading the lower and upper registers.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Move adis16201 driver out of staging

Move adis16201 driver out of staging and merge into mainline
IIO subsystem.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Fix 80 character line limit

Split the line over 80 characters limit to fix checkpatch
warning.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Use GENMASK

Use GENMASK to improve readability and remove the local variables used to
store intermediate data.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Remove unused headers

Remove few unused headers files since the adis core handles the buffer and
sysfs support.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: add copyright

Add Brian Masney's copyright and to the list of module authors for all
of the staging cleanups. This patch also update's Jon Brenner's current
work email address since AMS now owns TAOS.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: put local variables in reverse Christmas tree order

This patch ensures that all of the local variable declarations are in
reverse Christmas tree order where possible to increase code
readability.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: remove ch0 and ch1 variables from tsl2x7x_get_lux()

Remove the ch0 and ch1 variables from tsl2x7x_get_lux() and
write those values directly into the chip->als_cur_info.als_ch0
and chip->als_cur_info.als_ch01 variables.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: remove unused variables from tsl2x7x_get_lux()

tsl2x7x_get_lux() has a ch0lux and ch1lux variables that are not used
so this patch removes them.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: split out als and prox persistence settings

The struct tsl2x7x_settings contained a persistence member that
contained both the ALS and proximity persistence fields. This patch
splits this out into two separate fields so that the bitmasks in
several parts of the code are no longer necessary.

The default persistence settings are also changed by this patch from:

- Proximity: 0 (Every proximity cycle generates an interrupt)
- ALS: 255 (60 consecutive values out of range)

to something a little more reasonable based on my testing:

- Proximity: 1 (1 proximity value out of range)
- ALS: 1 (1 value outside of threshold range)

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: make logging consistent and correct newlines

This patch updates all of the logging commands so that they are
consistent with the other messages, includes __func__ in the message,
and all of the messages include newlines. This patch also removes some
debug log messages.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: split out als and prox interrupt settings

The struct tsl2x7x_settings contained an interrupts_en member that was
a bitmask for which interrupts are enabled. This required having
bitmasks in several parts of the code. This patch splits this field
out into two booleans to remove most of the bitmasks in the code.

This patch also fixes a bug where if an interrupt pin was configured,
but proximity interrupts were disabled, then the proximity value could
not be polled.

This patch also removes an unnecessary second call to writing the
control register in tsl2x7x_chip_on().

Driver tested using a TSL2772 hooked up to a Raspberry Pi 2.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: simplify tsl2x7x_prox_cal()

tsl2x7x_prox_cal() would set the interrupt flag, and reset the device to
start doing the calibration routine. However, this did not actually
affect the readings since they are polled. This patch drops the interrupt
code.

This patch also drops the function tsl2x7x_prox_calculate() and removes
support for the standard deviation and min sample since those values
were not used.

Driver was tested using a TSL2772 hooked up to a Raspberry Pi 2. I
performed the following testing at various distances:

- Put hand in front of sensor and keep the sensor and hand stationary.
- Perform calibration routine.
- Run iio_event_monitor.
- Verify that a proximity event is triggered when my hand comes
  anywhere between the sensor and where I performed the calibration
  routine.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: no need to clear interrupt flag when getting lux

tsl2x7x_get_lux() does not need to clear the interrupt flag when
querying the ALS. The interrupt flag is cleared in
tsl2x7x_event_handler(). This patches removes the unnecessary code.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: correct interrupt handler trigger

tsl2x7x_event_handler() was not called as expected when the device was
asserting a hardware interrupt. This patch changes the interrupt line
trigger from rising to falling.

The driver was tested on a TSL2772 hooked up to a Raspberry Pi 2. The
interrupt pin also had a 10K pull-up resistor per the requirements from
the datasheet. The relevant device tree binding:

&i2c1 {
tsl2772@39 {
compatible = "amstaos,tsl2772";
reg = <0x39>;
interrupt-parent = <&gpio>;
interrupts = <22 0x2>;
};
};

With this patch, iio_event_monitor now shows the events when the
channels are outside the defined interrupt thresholds.

$ sudo ./iio_event_monitor tsl2772
Found IIO device with name tsl2772 with device number 0
Event: time: 1478193460053760446, type: proximity, channel: 0, evtype:
thresh, direction: either
...
Event: time: 1478193463020270185, type: illuminance, channel: 0, evtype:
thresh, direction: either
...

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: remove unnecessary code

As a follow up to the work in commit a0722d05a195 ("staging: iio:
tsl2x7x: convert mutex_trylock() to mutex_lock()"), this patch removes
the unnecessary calls to tsl2x7x_get_prox() and tsl2x7x_get_lux() in
tsl2x7x_event_handler(). Previously, these functions were locked with
mutex_trylock(), but that is no longer the case. This patch also removes
a comment that is no longer relevant about returning the last sample.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ade7854: Remove read_reg_* duplications

The original code had a read function per data size; after updates, all
read functions tasks were centralized in a single function, but the old
signature was kept to maintain the module working without problems. This
patch removes a set of duplications associated with read_reg_*, and
update the areas that calling the old interface by the new one.

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ade7854: Rework SPI read function

Rework read SPI function to reduce the code duplication and centralizes
all the task in a single function.

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ade7854: Rework I2C read function

The read operation for the I2C function has many duplications that can
be generalized into a single function. This patch reworks the read
operation for I2C to centralizes all similar code in a single function.

It is possible to remove all the old interface to use the new one,
however, for keeping the things simple and working this patch maintain
legacy interface.

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ade7854: Remove write_reg_* duplications

This patch removes code duplications related to the write_reg_*
functions and centralizes them in a single function. Also, it eliminates
the legacy functions and replaces them by a unique signature that is
used by SPI and I2C.

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ade7854: Rework SPI write function

The write operation using SPI has a many code duplications (similar to
I2C) and four different interfaces per data size. This patch introduces
a single function that centralizes the main task related to SPI.

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ade7854: Rework I2C write function

The write operation using I2C has many code duplications and four
different interfaces per data size. This patch introduces a single
function that centralizes the main tasks.

The central function inserted by this patch can easily replace all the
four functions related to the data size. However, this patch does not
remove any code signature for keeping the meter module work and make
easier to review this patch.

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ade7854: Fix the wrong number of bits to read

Fixes: correctly handle the data size in the read operation for I2C

The function ade7854_i2c_read_reg_32() have to invoke the
i2c_master_recv() for read 32 bits values, however, the counter is set
to 3 which means 24 bits. This patch fixes the wrong size of 24 bits, to
32 bits.

Signed-off-by: John Syne <john3909@gmail.com>
Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Fixes: 8d97a5877 ("staging: iio: meter: new driver for ADE7754 devices")
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ade7854: Fix error handling on read/write

The original code does not correctly handle the error related to I2C
read and write. This patch fixes the error handling related to all
read/write functions for I2C.

Signed-off-by: John Syne <john3909@gmail.com>
Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Fixes: 8d97a5877 ("staging: iio: meter: new driver for ADE7754 devices")
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Adjust argument to match open parentheses

In adis16201_read_raw() adjust an argument to match an open parentheses
using tabs and spaces.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Prefer reverse christmas tree ordering

Prefer reverse christmas tree ordering of declarations to improve
readability.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Use sign_extend32 function

Use sign_extned32() for 32 bit sign extending rather than hard coding.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Add comments about units in read_raw()

Clarify the conversion and formation of resultant data in the
adis16201_read_raw() with sufficient comments and remove the unnecessary
comments.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Group register definitions

Group register definitions with its register field bits to improve
readability and easy identification. A small comment is also added to
denote the purpose/functionality of the grouped register definitions.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Add _REG suffix to registers

Add a _REG suffix to distinguish between registers and the register bit
fileds.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Remove unnecessary comments

Remove few unnecessary comments since the macro definitions clearly
justify their purpose.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: accel: adis16201: Rename few macro definitions

Rename the macro definitions with suitable names specifying their
purpose.

* ADIS16201_STARTUP_DELAY_MS: Remove the comment specifying the delay in
  microseconds and rename it with addtition of _MS suffix.

* ADIS16201_MSC_CTRL_ACTIVE_DATA_RDY_HIGH: Rename the macro to make it
  similar to other misc control registers which denotes the data ready
  polarity.

* ADIS16201_DIAG_STAT_FLASH_UPT_FAIL_BIT: Rename to denote it is a
  failure bit.

* ADIS16201_GLOB_CMD_FACTORY_RESET: Remove ambiguous _CAL suffix and add
  _RESET suffix instead to denote factory reset command.

Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: adc: ad7780: remove IIO_CHAN_INFO_SAMP_FREQ support

The `ad7780` driver does not implement setting/getting the sampling
frequency.
For the ad7780/ad7781 devices, the control is done via an external pin,
and the ad7170/ad7171 devices have a fixed sampling rate (so, no control).

For these devices, and similar other that may be added later on,
a AD_SD_CHANNEL_NO_SAMPLE_FREQ() macro has been added, which doesn't set
the IIO_CHAN_INFO_SAMP_FREQ flag.

Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: adc: ad7791: implement IIO_CHAN_INFO_SAMP_FREQ

Now that the old read/write frequency sysfs attrs have been removed, we
have a clean slate to implement IIO_CHAN_INFO_SAMP_FREQ.

This driver also pre-dates IIO_CHAN_INFO_SAMP_FREQ, and this change
implements this behavior.

The `ad7791_write_raw` would have overlapped quite a bit with the old
read/write frequency functions, making things a bit harder to follow.

Fixes: a13e831fcaa7 ("staging: iio: ad7192: implement
IIO_CHAN_INFO_SAMP_FREQ")

Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: adc: ad7791: remove sample freq sysfs attributes

In the current state, these attributes are broken, because they are
registered already, and the kernel throws a warning.
The first registration happens via the `IIO_CHAN_INFO_SAMP_FREQ` flag from
the `ad_sigma_delta` driver.

In this commit these attrs are removed, and in the following the
IIO_CHAN_INFO_SAMP_FREQ behavior will be implemented, which replaces these
hooks.

This is done to make things a bit easier to review as there is a bit of
overlap in the patch if it's done all at once.

Fixes: a13e831fcaa7 ("staging: iio: ad7192: implement IIO_CHAN_INFO_SAMP_FREQ")

Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: cros_ec: Move cros_ec_sensors_core.h in /include

Similar to other common iio frameworks, move cros_ec_sensors_core.h from
drivers/iio/common/cros_ec_sensors/ to include/linux/iio/common.

Signed-off-by: Gwendal Grignou <gwendal@chromium.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: stx104: Implement get_multiple callback

The Apex Embedded Systems STX104 series of devices provides 4 TTL
compatible lines of inputs accessed via a single 4-bit port. Since four
input lines are acquired on a single port input read, the STX104 GPIO
driver may improve multiple input reads by utilizing a get_multiple
callback. This patch implements the stx104_gpio_get_multiple function
which serves as the respective get_multiple callback.

Cc: Jonathan Cameron <jic23@kernel.org>
Cc: Hartmut Knaack <knaack.h@gmx.de>
Cc: Lars-Peter Clausen <lars@metafoo.de>
Cc: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
Signed-off-by: William Breathitt Gray <vilhelm.gray@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

uapi: remove telephony headers

ixjuser.h includes the telephony.h header. Other than that no kernel
code uses any of these headers. The last user of the ixjuser.h header
has been removed in commit 7326446c728 (Staging: remove telephony
drivers), more than 5 years ago.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-dpaa2/ethsw: Add TODO

Add a TODO file describing what needs to be added/changed before the driver
can be moved out of staging.

Signed-off-by: Razvan Stefanescu <razvan.stefanescu@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-dpaa2/ethsw: Add README

Add a README file describing the driver architecture, components and
interfaces.

Signed-off-by: Razvan Stefanescu <razvan.stefanescu@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-dpaa2/ethsw: Add maintainer for Ethernet Switch driver

Signed-off-by: Razvan Stefanescu <razvan.stefanescu@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-dpaa2/ethsw: Add ethtool support

Add driver information, link details and hardware statistics to be
reported via ethtool -S.

Signed-off-by: Razvan Stefanescu <razvan.stefanescu@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-dpaa2/ethsw: Add Freescale DPAA2 Ethernet Switch driver

Introduce the DPAA2 Ethernet Switch driver, which manages Datapath Switch
(DPSW) objects discovered on the MC bus.

Suggested-by: Alexandru Marginean <alexandru.marginean@nxp.com>
Signed-off-by: Razvan Stefanescu <razvan.stefanescu@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-dpaa2/ethsw: Add APIs for DPSW object

Add the command build/parse APIs for operating on DPSW objects through
the DPAA2 Management Complex.

Signed-off-by: Razvan Stefanescu <razvan.stefanescu@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-mc: Move DPCON out of staging

Move the source files out of staging into their final locations:
- dpcon.c goes to drivers/bus/fsl-mc/, next to the core infrastructure
- dpcon-cmd.h gets merged into drivers/bus/fsl-mc/fsl-mc-private.h, next
  to the other internally used APIs
- dpcon.h gets merged into include/linux/fsl/mc.h, exposing the public
  API

Update references in the dpaa2-eth staging driver.

DPCON stands for Data Path Concentrator - an interface between DPIO
(Data Path IO) and its users (e.g. dpaa2-eth). You can read more about
DPIO in Documentation/networking/dpaa2/overview.rst

Signed-off-by: Bogdan Purcareata <bogdan.purcareata@nxp.com>
Reviewed-by: Laurentiu Tudor <laurentiu.tudor@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-mc: Move DPBP out of staging

Move the source files out of staging into their final locations:
- dpbp.c goes to drivers/bus/fsl-mc/, next to the core infrastructure
- dpbp-cmd.h gets merged into drivers/bus/fsl-mc/fsl-mc-private.h, next
  to the other internally used APIs
- dpbp.h gets merged into include/linux/fsl/mc.h, exposing the public
  API

Update references in the dpaa2-eth staging driver.

DPBP stands for Data Path Buffer Pool - you can read more about the
object in Documentation/networking/dpaa2/overview.rst

Signed-off-by: Bogdan Purcareata <bogdan.purcareata@nxp.com>
Reviewed-by: Laurentiu Tudor <laurentiu.tudor@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: irda: remove the irda network stack and drivers

No one has publicly stepped up to maintain this broken codebase for
devices that no one uses anymore, so let's just drop the whole thing.

If someone really wants/needs it, we can revert this and they can fix
the code up to work properly.

Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: pi433: Remove comments inside code

Remove comments inbetween code as in this case.

Signed-off-by: HariPrasath Elango <hariprasath.elango@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:r8188eu: Use lib80211 to support TKIP

Custom TKIP decryption replaced with lib80211 implementation.
MIC check use lib80211 too.

Signed-off-by: Ivan Safonov <insafonov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Remove vchiq_queue_bulk_{transmit,receive}.

These are dead code, including in the downstream Raspberry Pi tree.

Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Mark the "DT bindings" job done.

Now we just need to get the other drivers merged and finish the style
cleanups/garbage collecting so we can get out of staging.

Signed-off-by: Eric Anholt <eric@anholt.net>
Acked-by: Stefan Wahren <stefan.wahren@i2se.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

ARM: dts: bcm2835: Add VCHIQ node to the Raspberry Pi boards. (v3)

The VCHIQ firmware communication channel operates in parallel with our
other mailbox-based channel.  This is the communication channel that
exposes the firmware's media decode/encode and ISP interfaces.

Signed-off-by: Eric Anholt <eric@anholt.net>
Acked-by: Stefan Wahren <stefan.wahren@i2se.com> (v2)
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

dt-bindings: soc: Add a binding for the Broadcom VCHIQ services. (v3)

The VCHIQ communication channel can be provided by BCM283x and Capri
SoCs, to communicate with the VPU-side OS services.

Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Remove cache-line-size property (v3)

It's been tempting to replace this with (L1) cache_line_size(), but
that's really not what the value is about.  It's about coordinating
the condition for the pagelist fragment behavior between the two
sides.  However, the property was not accepted for the upstream DT
binding, so we have to use the firmware's fallback value.

Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ccree: remove ccree staging driver copy

The ccree driver is now in the cryptodev tree, so remove it from
drivers/staging as it's no longer needed here.

Based on a patch from Gilad, but the mailing list didn't like it :(

Signed-off-by: Gilad Ben-Yossef <gilad@benyossef.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Replace "firmware" node with a compatible lookup.

This was requested by Rob Herring in DT bindings review.

Signed-off-by: Eric Anholt <eric@anholt.net>
Acked-by: Stefan Wahren <stefan.wahren@i2se.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: core: rtw_cmd: remove unnecessary initialization

Clean up checkpatch error:
ERROR: do not initialise globals to 0

Signed-off-by: Ji-Hun Kim <ji_hun.kim@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 char issue in handle_scan_done()

Fix 'line over 80 characters' issue found by checkpatch.pl script in
handle_scan_done().

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename variables using camelCase in handle_rcvd_gnrl_async_info()

Fix 'Avoid camelCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename WILC_HostIf_PackStaParam to avoid camelCase

Fix 'Avoid camelCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename variables prefix using datatype 'u8'

Rename variables with datatype 'u8' in their name to follow the linux
coding style.

Renamed following variables:
u8abort_running_scan
pu8Buffer
pu8keybuf
pu8msa
u8remain_on_chan_flag

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 character in handle_disconnect()

Refactor handle_disconnect() to avoid line over 80 characters issue
reported by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 char in handle_connect()

Fix 'line over 80 characters' issue reported by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 char in handle_scan()

Fix 'line over 80 character' issue reported by checkpatch.pl script in
handle_scan().

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename label 'ERRORHANDLER' to avoid uppercase name

Cleanup patch to avoid use of uppercase for label names, to follow linux
coding style.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename strConnectInfo to avoid camelCase

Fix 'Avoid camelCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename pstrHostIFconnectAttr to avoid camelCase issue

Fix 'Avoid camelCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 char in cfg_scan_result()

Refactor cfg_scan_result() API to avoid 'line over 80 chars' issue
reported by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename pJoinParams to avoid camelCase

Fix 'Avoid camelCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 char in mgmt_tx_cancel_wait()

Fix 'line over 80 char' issue found in checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 char for cfg parse RX and TX function

Fix 'line over 80 characters' issue found by checkpatch.pl script. Moved
the common code from wilc_wfi_cfg_parse_tx_action() &
wilc_wfi_cfg_parse_rx_action() to new function to avoid checkpatch issue.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 char in get_key() & set_default_key()

Fix 'line over 80 characters' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: fix line over 80 char in change_virtual_intf()

Fix 'line over 80 char' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename au8ScanChanList to avoid camelCase

Fix 'Avoid camelCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename CfgScanResult to avoid camelCase

Fix 'Avoid camelCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename pstrNetworkInfo to avoid camelCase

Fix 'Avoid camleCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename strHiddenNetwork to avoid camelCase

Fix 'Avoid camelCase' issue found by checkpatch.pl script.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: Add unicode support to the speakup_dummy driver

This extends spk_io_ops with a synth_out_unicode which takes a u16 character
instead of just a byte, and extends spk_ttyio to implement it to emit
utf-8. spk_do_catch_up_unicode can then be introduced to benefit from
synth_out_unicode, and speakup_dummy made to use spk_do_catch_up_unicode instead
of spk_do_catch_up.

Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: adl_pci6208: remove redundant initialization of 'val'

Variable 'val' is initialized with a value that is never read, it is
updated with a new value again after intitialization. Remove the
redundant initialization and move the declaration and assignment into
the scope of the for-loop.

Cleans up clang warning:
drivers/staging/comedi/drivers/adl_pci6208.c:61:15: warning: Value
stored to 'val' during its initialization is never read

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ks7010: replace custom rotation operations in favour of the kernel ones

This patch replaces custom ROR32 and ROL32 macros for the ones included in
bitops header of the linux kernel.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: fid: avoid false-positive uninitialized variable warning

One of Neil's recent cleanups apparently has led the code to get
to a state where gcc tracks the 'seqnr' variable just enough to
see that it is sometimes initialized in seq_client_alloc_seq(),
but not enough that it can prove this initialization to be reliable
before the use of that variable:

drivers/staging/lustre/lustre/fid/fid_request.c: In function 'seq_client_alloc_fid':
drivers/staging/lustre/lustre/fid/fid_request.c:245:22: error: 'seqnr' may be used uninitialized in this function [-Werror=maybe-uninitialized]

The code seems to be otherwise correct, and I could not come
up with a good way to simplify it further, so this adds a fake
initialization to shut up that warning.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: use module_name() macro for debug printf

We now allow lustre to be built when CONFIG_MODULES is disabled,
but that causes a build failure:

In file included from drivers/staging/lustre/include/linux/libcfs/libcfs.h:42,
                 from drivers/staging/lustre/lustre/obdclass/lu_object.c:44:
drivers/staging/lustre/lustre/obdclass/lu_object.c: In function 'lu_context_key_degister':
drivers/staging/lustre/lustre/obdclass/lu_object.c:1410:51: error: dereferencing pointer to incomplete type 'struct module'

This particular case can be avoided by using the module_name()
macro that was designed exactly to handle printing the name of
a module in all configurations.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge tag 'iio-for-4.17b' of git://git./linux/kernel/git/jic23/iio into staging-next

Jonathan writes:

Second set of new device support, features and cleanup for IIO in the 4.17 cycle

The uptick in staging cleanup is partly due to GSoC Applications
process being underway and one of Daniel's tasks being to try
cleaning up an IIO driver to move out of staging.
Naturally there is some normal staging cleanup progress in here
as well.

New device support
* Microchip mcp4018
  - New driver supproting MCP4017, MCP4018 and MCP4019 digital pots.
* On Semiconductor lv0104cs
  - New driver to support this ambient light sensor.

Cleanup
* axp20x_adc
  - remove a !! in favour of clear ternary operator.
* ad2s1210 (staging cleanup)
  - Reorganise to avoid ending a line with [
  - Remove some unnecessary defines.
  - Remove unsed variable.
* ad5380
  - Replace magic 0 with IIO_CHAN_INFO_RAW
* ad5764
  - Replace magic 0 with IIO_CHA_INFO_RAW
* ad7150 (staging cleanup)
  - Align arguements with parenthesis.
* ad7152 (staging cleanup)
  - Align arguements.
* ad7746 (staging cleanup)
  - Align arguements.
* ad7816
  - Remove pointless void pointer cast.
* ade7753
  - Replace IIO_DEV_ATTR_CH_OFF with equivalent to avoid confusing
    checkpatch (this macro didn't really help anyway).  Also drop the
    macro from the meter.h header.
* ade7754 (staging cleanup)
  - Add names to funciton definition arguements.
  - Align arguements with open parenthesis where possible.
* ade7758 (staging cleanup)
  - Remove __func__ from dev_dbg statements as provided by dynamic
    debug anyway.
  - Align arguements with open parenthesis where possible.
* ade7759 (staging cleanup)
  - Replace IIO_DEV_ATTR_CH_OFF with equiavalent to avoid confusing
    checkpatch.
* adis16201 (staging cleanup)
  - Headers in alphabetical order.
  - Blank lines before returns.
* adis16209 (staging cleanup)
  - Headers in alphabetical order
  - Change some definition names to make them more meaningful (2 rounds
    of this).
  - Add explicit _REG prefix to register names to make them
    obviously different from fields within those registers.
  - Remove some superflous comments and group definitions better.
  - Use a switch statement to make it semantically obvious that we
    only have two options (rather than an unlimited 'else').
  - Use sign_extent32 instead of open coding.
* adt7316 (staging cleanup)
  - Move an export next to symbol.
* bmc150
  - drop redundant __func__ in dynamic debug.
* ccs811
  - Rename varaible to better reflect what it does.
* cros_ec
  - Reduce sampling frequency before suspending to avoid preventing
    suspend.
* dummy
  - Correct whitespace in Kconfig.
  - Add extra description in Kconfig.
* ds1803
  - Remove a VLA which we always know is 2 long.
* hid-sensor-accel
  - Replace magic number 0 by IIO_CHAN_INFO_RAW.
* hid-sensor-gyro
  - Replace magic number 0 by IIO_CHAN_INFO_RAW.
* hid-sensor-light
  - Replace magic number 0 by IIO_CHAN_INFO_RAW.
* hid-sensor-magn
  - Replace magic number 0 by IIO_CHAN_INFO_RAW.
* lm3533
  - Replace magic number 0 by IIO_CHAN_INFO_RAW
* mlx90632
  - Squash a smatch warning - no runtime effect.
* stm32_dfsdm:
  - Cleanup the dt bindings.
* sx9500
  - Add GPIO ACPI mapping table to behave correctly when firmware
    doesn't provide the mapping.
* tsl2x7x (staging cleanup)
  - Fix the proximity sensor functionality.
  - Remove platform data provided power functions.  There are much
    better ways to do this these days.
  - Introduce some common functions to avoid various repititions.
  - Stop using mutex_trylock when mutex_lock and wait a bit is fine.
  - Improve error handling in various places.
  - Drop some 'Camel case' (which wasn't actually strickly camel case
    but was a bit odd.
  - Drop some _available sysfs attributes for things that don't exist
    (for particular supported parts).

staging: fsl-dpaa2/eth: allow the driver to compile multi-arch

Drop dependency on ARCH_LAYERSCAPE (which in turn depends on ARM64),
thus allowing this driver to compile on all architectures supported
by the fsl-mc bus driver.

This was compile tested on:
 - powerpc (corenet_basic_defconfig, ppc64_defconfig)
 - x86 (i386_defconfig, x86_64_defconfig, needs CONFIG_OF)
 - arm64 (defconfig)

Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fsl-dpaa2/eth: Use __leXX types where needed

One MC command structure got away with using uXX fields instead
of __leXX. Fix it.

Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:iio:ad2s1210: Remove old_data from ad2s1210_state

The variable old_data is a bool type, which only receives the value
'true' in the function ad2s1210_config_write and ad2s1210_config_read.
There is no other use for this variable. This patch removes old_data
from the ad2s1210_state and from all the function that use it.

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ad2s1210: Remove unused #define directive

This patch removes some #define directives not used in the code.

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ad2s1210: Remove end of line with '['

This patch fixes the checkpatch.pl check:

iio/resolver/ad2s1210.c:202: CHECK: Lines should not end with a '['

Signed-off-by: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ad7150: Align arguments to match open parenthesis

This fixes the following checkpatch.pl CHECK: Alignment should match
open parenthesis

Signed-off-by: Tom Lebreux <tomlebreux@cock.li>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging:iio:ad7152: Align arguments to match open parenthesis

This patch fixes checkpatch.pl CHECK for alignment.

Signed-off-by: Tom Lebreux <tomlebreux@cock.li>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: potentiometer: mcp4018: driver for Microchip digital potentiometers

Add support for Microchip digital potentiometers and rheostats
MCP4017, MCP4018, MCP4019

They all have one wiper with 128 steps and come in 5, 10, 50 and 100 kOhm
variations.

Datasheet: http://www.microchip.com/downloads/en/DeviceDoc/22147a.pdf

Signed-off-by: Peter Rosin <peda@axentia.se>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: ad7746: Adjust arguments to match open parenthesis

This patch clears all checkpatch.pl CHECKS, about alignment not matching
open parenthesis, whenever it is possible without going beyond 80 columns.

Signed-off-by: David Veenstra <davidjulianveenstra@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: adt7316: Move symbol export to definition

This patch clears the following checkpatch.pl warning:

WARNING: EXPORT_SYMBOL(foo); should immediately follow its function/variable
+EXPORT_SYMBOL_GPL(adt7316_pm_ops);

Signed-off-by: David Veenstra <davidjulianveenstra@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: potentiometer: ds1803: Remove VLA usage

In preparation to enabling -Wvla, remove VLA usage and replace it
with fixed a fixed length array and therefore, prevent potential
stack overflow attacks.

Fixed as a part of the discussion to remove all VLAs from the kernel:
https://lkml.org/lkml/2018/3/7/621

Cc: keescook@chromium.org
Signed-off-by: Himanshu Jha <himanshujha199640@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: make proximity sensor function correctly

The bits for setting up the proximity diode were not setup correctly and
this was causing the proximity sensor to not function correctly. This
patch sets up the correct bit mask in tsl2x7x_chip_on() based on what
the data sheet expects. From page 35 of the TSL2771 data sheet:
https://ams.com/eng/content/download/250264/976045/file/TSL2771_DS000105_3-00.pdf

- Bits 0-1 is the ALS gain control
- Bits 2-3 is reserved (The proximity gain control on other tsl2x7x chips)
- Bits 4-5 is the proximity diode select
- Bits 6-7 is the LED drive strength

tsl2x7x_chip_on() had the power and diode hardcoded, so these are
extracted out into the settings so that these fields can be configured
in the platform data.

The default prox_gain is changed from 1 (2X gain) to 0 (1X gain) since
the proximity gain control on the TSL2771, TMD2771, and other chips have
these fields listed as reserved, and to write 0 into those bits.

Verified that the proximity sensor now works correctly on a TSL2771
hooked up to a Raspberry Pi 2.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: remove unnecessary sysfs attribute

The tsl2771 and tmd2771 devices create the
in_proximity0_calibscale_available sysfs attribute. These two particular
devices do not support changing the proximity gain value on the
chip so this patch removes that sysfs attribute. As expected, these two
devices already did not create the IIO_CHAN_INFO_CALIBSCALE channel and
proximity0_calibrate sysfs attribute.

Page 38 of the tsl2772 data sheet shows that the proximity gain can be
adjusted with bits 2-3 on the control register:
https://ams.com/eng/content/download/291503/1066377/file/TSL2772_DS000181_2-00.pdf

Page 35 of the tsl2771 and tmd2771 data sheets shows that bits 2-3 on
the control register are reserved and changing the proximity gain is
not supported:
https://ams.com/eng/content/download/250264/976045/file/TSL2771_DS000105_3-00.pdf
https://ams.com/eng/content/download/250283/976077/file/TMD2771_DS000177_2-00.pdf

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: add missing error checks

The functions in_illuminance0_calibrate_store() and
in_illuminance0_lux_table_store() did not have complete error handling
in place. This patch adds the missing error handling.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: add error handling to tsl2x7x_prox_cal()

tsl2x7x_prox_cal() did not have any error checks. This patch adds
the missing error handling and ensures that any errors are reported to
user space via in_proximity0_calibrate_store().

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: correct 'Avoid CamelCase' warning from checkpatch

The statP and calP variables triggered an 'Avoid CamelCase' warning
from checkpatch.pl. This patch renames these variables to stat and cal
to fix this warning.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: correctly return errors in tsl2x7x_get_prox()

Not all errors that occurred in tsl2x7x_get_prox() were correctly
reported in the return value. This patch changes the error handling
so that errors are now returned properly.

Note that the ret variable is from the call to tsl2x7x_read_status(),
and it already has the correct error check. The -EINVAL error code is
for an unexpected value in the register.

This patch also corrects an unnecessary word wrap in the call to
le16_to_cpup() while changes are being made to this code.

Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>