Select IIO_BUFFER and IIO_TRIGGERED_BUFFER to compile.
Remove IIO_TRIGGER if present since IIO_BUFFER selects it.
Signed-off-by: NAlison Schofield <amsfield22@gmail.com>
Cc: Daniel Baluta <daniel.baluta@gmail.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Added macros for sensing range as the corresponding magic numbers
were used at multiple places.
   - ISL29125_SENSING_RANGE_0 for 375 lux full range
   - ISL29125_SENSING_RANGE_1 for 10k lux full range
Signed-off-by: NBijosh Thykkoottathil <bijosh.t@hotmail.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Adds a new per-device sysfs attribute "current_timestamp_clock" to allow
userspace to select a particular POSIX clock for buffered samples and
events timestamping.

Following clocks, as listed in clock_gettime(2), are supported:
CLOCK_REALTIME, CLOCK_MONOTONIC, CLOCK_MONOTONIC_RAW,
CLOCK_REALTIME_COARSE, CLOCK_MONOTONIC_COARSE, CLOCK_BOOTTIME and
CLOCK_TAI.
Signed-off-by: NGregor Boirie <gregor.boirie@parrot.com>
Acked-by: NSanchayan Maity <maitysanchayan@gmail.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Replace the code that guarantees the device stays in direct mode
with iio_device_claim_direct_mode() which does same.  Includes a
tiny bit of refactoring (single case -> if) and simplified return
flow.

Also prevents a theoretical race condition by effectively taking the lock
a little earlier than before.
Signed-off-by: NAlison Schofield <amsfield22@gmail.com>
Cc: Daniel Baluta <daniel.baluta@gmail.com>
Acked-by: NJacek Anaszewski <j.anaszewski@samsung.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Replace the code that guarantees the device stays in direct mode
with iio_device_claim_direct_mode() which does same.  This allows
removal of an unused lock in the device private global data.
Signed-off-by: NAlison Schofield <amsfield22@gmail.com>
Cc: Daniel Baluta <daniel.baluta@gmail.com>
Acked-by: NPeter Meerwald-Stadler <pmeerw@pmeerw.net>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Replace the code that guarantees the device stays in direct mode
with iio_device_claim_direct_mode() which does same.  This allows
removal of an unused lock in the device private global data.

Also prevents a previous race condition when a buffer is enabled whilst
a raw read is going on.
Signed-off-by: NAlison Schofield <amsfield22@gmail.com>
Cc: Daniel Baluta <daniel.baluta@gmail.com>
Acked-by: NPeter Meerwald-Stadler <pmeerw@pmeerw.net>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Replace the code that guarantees the device stays in direct mode
with iio_device_claim_direct_mode() which does same.  This allows
removal of an unused lock in the device private global data.

Also prevents a possible race around a buffer being enabled whilst
a read is going on.
Signed-off-by: NAlison Schofield <amsfield22@gmail.com>
Cc: Daniel Baluta <daniel.baluta@gmail.com>
Acked-by: NPeter Meerwald-Stadler <pmeerw@pmeerw.net>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Without this, the iio:deviceX is missing in the /sys/bus/i2c/devices/0-0039
Some userspace tools use this path to identify a specific instance of the
device.
Signed-off-by: NYong Li <sdliyong@gmail.com>
Reviewed-By: NMatt Ranostay <mranostay@gmail.com>
Cc: <Stable@vger.kernel.org>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Using the struct i2c_device->id field for naming the light sensor
is a bad idea: when booting from the pure device tree this is NULL
and that causes the device not to have the "name" property in
sysfs and that in turn confuses the "lsiio" command to stop listing
devices.

So instead of using the device .id, use the hard string "bh1780",
which works just fine.

Fixes: 1f0477f1 ("iio: light: new driver for the ROHM BH1780")
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Cc: <Stable@vger.kernel.org>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

The code in runtime_[suspend|resume] was assuming that the
i2c client data was the bh1780 state container, but it contains
the IIO device. So first dereference the IIO device from the
i2c client, then get the state container using the iio_priv()
call.

Fixes: 1f0477f1 ("iio: light: new driver for the ROHM BH1780")
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Cc: <Stable@vger.kernel.org>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

This is not implemented and doesn't really make sense because IIO
proximity is unit-less.

Remove IIO_CHAN_INFO_SCALE from info_mask because so that the _scale
sysfs entry won't appear. This fixes userspace tools like generic_buffer
which abort when reads returns an error.
Signed-off-by: NCrestez Dan Leonard <leonard.crestez@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

This driver does not call i2c_smbus_read|write_byte_data(),
so remove the corresponding functionality test. It uses regmap
to handle byte transfers transparently.
Signed-off-by: NAlison Schofield <amsfield22@gmail.com>
Reviewed-by: NKuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Reviewed-by: NMatt Ranostay <matt.ranostay@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

When writing a value using direct reg access from debugfs
we need to return and not fall through to reading the
value, lest we'll dereference a NULL pointer.

Cc: Dan Carpenter <dan.carpenter@oracle.com>
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Signed-off-by: NCrestez Dan Leonard <leonard.crestez@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

This patch exposes ALSTIM as illuminance_integration_time and ALSPGA as
illuminance_scale.

Changing ALSTIM also changes the number of bits available in the data
register. This is handled inside raw value reading because:
* It's very easy to shift a few bits
* It allows SCALE and INT_TIME to be completely independent controls
* Buffer support requires constant scan_type.realbits per-channel
Signed-off-by: NCrestez Dan Leonard <leonard.crestez@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

This is exposed as an output channel with "led" as an extend_name.

Other sensors also have support for controlling an external LED. It's
not clear that simply exposing an undecorated output channel is the
correct approach.
Signed-off-by: NCrestez Dan Leonard <leonard.crestez@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

The proximity sensor relies on sending pulses to an external IR led and
it is disabled by default on powerup. The driver will enable it with a
default power setting.
Signed-off-by: NCrestez Dan Leonard <leonard.crestez@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

This just adds support for reporting illuminance with default settings.

Important default registers are written on probe because the device
otherwise lacks a reset function.
Signed-off-by: NCrestez Dan Leonard <leonard.crestez@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

It causes a static checker warning if we use "buf" on the failure path
so move that inside the if statement.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

ultraviolet (UV) light sensor with I2C interface with a peak
sensitivity at 355 nm

strangely, chip uses two addresses 0x38 and 0x39 for LSB and
MSB data, resp.

datasheet: http://www.vishay.com/docs/84277/veml6070.pdfSigned-off-by: NPeter Meerwald-Stadler <pmeerw@pmeerw.net>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

This is a reimplementation of the old misc device driver for the
ROHM BH1780 ambient light sensor (drivers/misc/bh1780gli.c).

Differences from the old driver:
- Uses the IIO framework
- Uses runtime PM to idle the hardware after 5 seconds
- No weird custom power management from userspace
- No homebrewn values in sysfs

This uses the same (undocumented) device tree compatible-string
as the old driver ("rohm,bh1780gli").

Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Ulf Hansson <ulf.hansson@linaro.org>
Cc: Daniel Mack <daniel@caiaq.de>
Cc: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Reviewed-by: NUlf Hansson <ulf.hansson@linaro.org>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

This fixes the error reported by checkpatch.pl:

ERROR: that open brace { should be on the previous line
Signed-off-by: NSlawomir Stepien <sst@poczta.fm>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

GPIO handling code has been removed from the drivers (since
this is now handled by the ACPI core) in commit 0f079650 ("iio:
remove gpio interrupt probing from drivers that use a single interrupt").

Remove the include for linux/gpio/consumer.h since it is no longer
used.
Signed-off-by: NIrina Tirdea <irina.tirdea@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

flush_scheduled_work is scheduled for deprecation.
Replace cancel_delayed_work and flush_scheduled_work with
cancel_delayed_work_sync instead to ensure there is no pending or
running work item.

Since there is only one work item, chip->poweroff_work, there are
no further dependencies of flush_scheduled_work().
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NAmitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Correct issue that the last entry in FIFO was being read twice due
to an incorrect decrement of entry count variable before condition
check.
Signed-off-by: NMatt Ranostay <mranostay@gmail.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Previously most drivers that used a i2c_check_functionality() check
condition required various error codes on failure. This patchset
converts to a standard of -EOPNOTSUPP
Signed-off-by: NMatt Ranostay <mranostay@gmail.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Enable operation of the TI OPT3001 light sensor without having an
interrupt line available to connect the INT pin to.

In this operation mode, we issue a conversion request and simply wait
for the conversion time available as timeout value, determined from
integration time configuration and the worst-case time given in the data
sheet (sect. 6.5, table on p. 5):

  short integration time (100ms): 110ms + 3ms = 113ms
   long integration time (800ms): 880ms + 3ms = 883ms

This change is transparent as behaviour defaults to using the interrupt
method if an interrupt no. is configured via device tree. Interrupt-less
operation mode is performed when no valid interrupt no. is given.
Signed-off-by: NAlexander Koch <mail@alexanderkoch.net>
Signed-off-by: NMichael Hornung <mhornung.linux@gmail.com>
Tested-by: NAndreas Dannenberg <dannenberg@ti.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Change variable type of struct opt3001 members 'ok_to_ignore_lock' and
'result_ready' uint16-bitfield of length one to bool.

They are used as bool, let the compiler do the optimization.
Signed-off-by: NAlexander Koch <mail@alexanderkoch.net>
Signed-off-by: NMichael Hornung <mhornung.linux@gmail.com>
Tested-by: NAndreas Dannenberg <dannenberg@ti.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Extract integration times as #define constants. This prepares using them
for delay/timeout length determination.
Signed-off-by: NAlexander Koch <mail@alexanderkoch.net>
Signed-off-by: NMichael Hornung <mhornung.linux@gmail.com>
Tested-by: NAndreas Dannenberg <dannenberg@ti.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

As per the ACPI specification (Revision 5.0) [1], the data coming
from the sensor represent the ambient light illuminance reading
expressed in lux. So use IIO_CHAN_INFO_PROCESSED to signify that
the data are pre-processed.

However, to keep backward ABI compatibility, the IIO_CHAN_INFO_RAW
bit is not removed.

[1] http://www.acpi.info/DOWNLOADS/ACPIspec50.pdf

This issue has also been responsible for at least one userspace bug
report hence marking what is a small semantic fix really for stable.
[2] https://github.com/hadess/iio-sensor-proxy/issues/46Signed-off-by: NGabriele Mazzotta <gabriele.mzt@gmail.com>
Cc: <Stable@vger.kernel.org>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

A bit of refactoring for better readability.
Moved and slightly reorganized all the activity necessary for reading als
and proximity into a different function. This way the switch in read raw
becomes clearer and more compact.
Signed-off-by: NAdriana Reus <adriana.reus@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Add interrupt support for proximity.
Add two threshold events to signal rising and falling directions.
Signed-off-by: NAdriana Reus <adriana.reus@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

When setting als only or proximity only modes make sure that we mark the
other component as disabled. This fix is in preparation of adding event
support because that will make it possible to switch between one-shot and
continuous modes and not tracking these correctly may cause faulty
behaviour (e.g wrongfully considering px enabled and not setting an
appropriate mode in the chip).
Signed-off-by: NAdriana Reus <adriana.reus@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

The return type "unsigned int" was used by the ltr501_match_samp_freq()
function despite of the aspect that it will eventually return a negative
error code.
Improve this implementation detail by deletion of the type modifier then.

This issue was detected by using the Coccinelle software.
Signed-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Acked-by: NPeter Meerwald-Stadler <pmeerw@pmeerw.net>
Cc: <Stable@vger.kernel.org>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Add power management for sleep as well as runtime pm.
Signed-off-by: NAdriana Reus <adriana.reus@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Keep track of the als and px enabled/disabled status in
order to enable them selectively.
Signed-off-by: NAdriana Reus <adriana.reus@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

This chip supports two power modes.
1. "one-shot" mode - the chip activates and executes one complete
conversion loop and then shuts itself down. This is the default mode
chosen for raw reads.
2. "continuous" mode - the chip takes continuous measurements.

Continuous mode is more expensive power-wise but may be more reliable.
Add a property so that if preferred, the default power mode for raw
reads can be set to continuous.
Separate one-shot enabling in a separate function that will be used
depending on the chosen power mode. Also create a function for
powering the chip on and off.
Signed-off-by: NAdriana Reus <adriana.reus@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Add missing pm_runtime_mark_last_busy to apds9960_set_power_state
function.

Unless pm_runtime_mark_last_busy is called the
pm_runtime_put_autosuspend may put the device into suspend before the
delay time requested.
Signed-off-by: NMatt Ranostay <mranostay@gmail.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

At probe, runtime pm should be setup before registering the sysfs interface so
that all the power attributes are accurate and functional when registering.
Also, when removing the device we should unregister first to make sure
that the interfaces that may result in wakeups are no longer available.

Fix this behaviour for the following drivers: bmc150, bmg160, kmx61,
kxcj-1013, mma9551, mma9553, rpr0521.
Signed-off-by: NAdriana Reus <adriana.reus@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Make sure we poweroff the chip if for any reason iio_register
returns an error.
Signed-off-by: NAdriana Reus <adriana.reus@intel.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>