commit 95ad67577d upstream.
iio_push_to_buffers_with_timestamp assumes 8 byte alignment which
is not guaranteed by an array of smaller elements.
Note that whilst in this particular case the alignment forcing
of the ts element is not strictly necessary it acts as good
documentation. Doing this where not necessary should cut
down on the number of cut and paste introduced errors elsewhere.
Fixes: 0427a106a9 ("iio: accel: kxsd9: Add triggered buffer handling")
Reported-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Based on 2 normalized pattern(s):
this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license version 2 as
published by the free software foundation
this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license version 2 as
published by the free software foundation #
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-only
has been chosen to replace the boilerplate/reference in 4122 file(s).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Enrico Weigelt <info@metux.net>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190604081206.933168790@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently mount matrix is allowed in Device Tree, though there is
no technical issue to extend it to support ACPI.
Convert the function to use device_property_read_string_array() and
thus allow to read mount matrix from ACPI if available.
Example of use in _DSD method:
Name (_DSD, Package ()
{
ToUUID("daffd814-6eba-4d8c-8a91-bc9bbf4aa301"),
Package ()
{
Package () { "mount-matrix", Package() {
"1", "0", "0",
"0", "0.866", "0.5",
"0", "-0.5", "0.866",
} },
}
})
At the same time drop the "of" prefix from its name and
convert current users.
No functional change intended.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The equivalent of both of these are now done via macro magic when
the relevant register calls are made. The actual structure
elements will shortly go away.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Lars-Peter Clausen <lars@metafoo.de>
This adds support for the mounting matrix to the KXSD9 driver.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
This deploys runtime and system PM in the KXSD9 driver:
- Use the force_runtime_suspend/resume callbacks as system PM
operations.
- Add buffer prepare/unprepare callbacks to grab the runtime
PM while we're using buffered reads and put get/put_autosuspend
in these.
- Insert get/put_autosuspend calls anywhere the IO is used from
the raw read/write callbacks.
- Move the fullscale setting to be cached in the state container
so we can restore it properly when coming back from
system/runtime suspend.
- Set the autosuspend delay to two orders of magnitude that of
the sensor start-up time (20ms) so we will autosuspend after
2s.
- Register the callbacks in both the SPI and I2C subdrivers.
Tested with the I2C KXSD9 on the Qualcomm APQ8060 Dragonboard.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
What is passed to the .probe() and .remove() functions is
technically the parent of the created IIO device but it becomes
a big confusion for the head to have it named like this since
it is usually clear from context the "dev" refers to the physical
device, and when next adding PM callbacks a clean
"struct device *dev" pointer is passed to these and that makes
it even more confused. Rename "parent" to "dev" like in most
other drivers.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
This adds supply regulator handling for the VDD and IOVDD inputs
on the KXSD9 component, makes sure to bring the regulators online
during probe and disable them on remove or the errorpath.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
There are some hardcoded register values etc in the code, define
proper bitfield definitions, and use them when getting and setting
the scale. Optimize a read/modify/write to use regmap_update_bits()
at the same time.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
As is custom with all modern sensors, add a clever burst mode
that will just stream out values from the sensor and provide it
to userspace to do the proper offsetting and scaling.
This is the result when tested with an HRTimer trigger:
$ generic_buffer -a -c 10 -n kxsd9 -t foo
/sys/bus/iio/devices/iio:device1 foo
0.371318 0.718680 9.869872 1795.000000 97545896129
-0.586922 0.179670 9.378775 2398.000000 97555864721
-0.299450 0.179670 10.348992 2672.000000 97565874055
0.371318 0.335384 11.103606 2816.000000 97575883240
0.179670 0.574944 10.540640 2847.000000 97585862351
0.335384 0.754614 9.953718 2840.000000 97595872425
0.179670 0.754614 10.732288 2879.000000 97605882351
0.000000 0.754614 10.348992 2872.000000 97615891832
-0.730658 0.574944 9.570422 2831.000000 97625871536
0.000000 1.137910 10.732288 2872.000000 97635881610
Columns shown are x, y, z acceleration, so a positive acceleration
of ~9.81 (shaky due to bad calibration) along the z axis. The
fourth column is the AUX IN which is floating on this system,
it seems to float up to the 2.85V VDD voltage.
To be able to cleanup the triggered buffer, we need to add .remove()
callbacks to the I2C and SPI subdrivers and call back into an
exported .remove() callback in the core.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
This fixes several errors in the offset and scaling of the raw
values from the KXSD9 sensor:
- The code did not convert the big endian value from the sensor
into the endianness of the host CPU. Fix this with
be16_to_cpu() on the raw obtained value.
- The code did not regard the fact that only the upper 12 bits of
the accelerometer values are valid. Shift these
down four bits to yield the real raw value.
- Further the sensor provides 2048 at zero g. This means that an
offset of 2048 must be subtracted from the raw value before
scaling. This was not taken into account by the driver,
yielding a weird value. Fix this by providing this offset in
sysfs.
To house the scaling code better, the value reading code was
factored into the raw reading function.
This proper scaling and offseting is necessary to get proper
values out of triggered buffer by offsetting, shifting and scaling
them.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The RX/TX buffers are gone so drop the lock (it should have been
in the transport struct anyway).
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
This converts the KXSD9 driver to drop the custom transport
mechanism and just use regmap like everything else.
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
This is just a masquerading register write function, so use the
register write function instead.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
This moves the KXSD9 SPI transport out to its own file and Kconfig
entry, so that we will be able to add another transport method.
We export the common probe and add a local header file for the
functionality shared between the main driver and the transport
driver.
We make the SPI transport the default for the driver if SPI is
available and the KXSD9 driver was selected, so the oldconfig
upgrade path will be clear.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
This makes it possible to later split the transport mechanism
using a generic probe() and a generic remove().
Use dev_set_drvdata() and dev_get_drvdata() as a paired
accessor to operate on the abstract struct device * regardless
of the transport mechanism in use.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Split off a transport mechanism struct that will deal with the SPI
traffic in preparation for adding I2C support.
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
All the scaling of the KXSD9 involves multiplication with a
fraction number < 1.
However the scaling value returned from IIO_INFO_SCALE was
unpredictable as only the micros of the value was assigned, and
not the integer part, resulting in scaling like this:
$cat in_accel_scale
-1057462640.011978
Fix this by assigning zero to the integer part.
Cc: stable@vger.kernel.org
Tested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Any readings from the raw interface of the KXSD9 driver will
return an empty string, because it does not return
IIO_VAL_INT but rather some random value from the accelerometer
to the caller.
Cc: stable@vger.kernel.org
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
These two spi_w8r8() calls return a value with is used by the code
following the error check. The dubious use was caused by a cleanup
patch.
Fixes: d34dbee8ac ("staging:iio:accel:kxsd9 cleanup and conversion to iio_chan_spec.")
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
An spi_driver does not need to set an owner, it will be populated by the
driver core.
Signed-off-by: Andrew F. Davis <afd@ti.com>
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
This will leave a lock held after reading from the device, preventing
any further reads.
Signed-off-by: Frank Zago <frank@zago.net>
Cc: stable@vger.kernel.org
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Using devm_iio_device_alloc makes code simpler.
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Use the new spi_sync_transfer() helper function instead of open-coding it.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Jonathan Cameron