The default decoder would take the micro-unit value of the old sensor
value and multiply it by INT32_MAX. This would, at times, cause an
overflow for the int64_t which is the cause of some bugs like when
-7952 was used (-7952000000 * INT32_MAX < INT64_MIN). Instead the new
math converts:
- `value_u * INT32_MAX / ((1 << header->shift) * 1000000)`
to a bitmap:
- `sample.val1` consumes the upper `N` bits
- `sample.val2 * BIT(32 - N) / 1000000` consumes the lower `32-N`
bits
This both improves the accuracy, and avoids the overflow since
`shift` is guaranteed to be between 0 and 31.
Signed-off-by: Yuval Peress <peress@google.com>
Implement the backend emul API for the ICM42688 motion sensor so it can
be automatically tested by the generic sensor test (see #60394).
Supports all channels (temp, accel XYZ, and gyro XYZ) at each of the
programmable full-scale accel and gyro ranges.
Also fixes an arithmetic bug in the driver that was causing a minor
error in the returned readings.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
Add support for fetching individual channels rather than forcing the
application to always fetch them all. Potentially saves few i2c
transactions.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Convert a few floating point operation to fixed point. May save some
stack and some flash as well on soft-fp systems.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
The register space for the device is on 8 bit, use uint8_t for register
address to the low level read and write functions.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
sensor_value_from_double was used, but sensor_value_from_float
should be used as the parameter is a float
Signed-off-by: Ryan McClelland <ryanmcclelland@meta.com>
The LIS3MDL driver misunderstands the MD[1:0] bits in CTRL_REG3.
Correct this so that all ODR settings are functional.
Currently, single-conversion mode is enabled when fast ODR is set,
resulting in only one measurement being taken before the device
immediately returns to power-down mode. The result is that all fast-ODR
rates (155 and higher) are not usable when assigned to CONFIG_LIS3MDL_ODR.
Continuous mode is the only mode truly supported by this driver and should
always be used.
Signed-off-by: Aedan Cullen <aedan@aedancullen.com>
Since the sensor shell command was converted to use qt31_t, all the
integer values started to show up as rounded up by a fractional unit
when displayed, due to the conversion always rounding down.
Fix that by using the recently introduced DIV_ROUND_CLOSEST and handling
rounding up to next integer explicitly.
Before:
channel idx=44 gauge_state_of_charge value=83.999999
after:
channel idx=44 gauge_state_of_charge value=84.000000
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Format the file `drivers/sensor/max17262/max17262.c` but exclude the
regsiter lookup table in `max17262_sample_fetch()`
Signed-off-by: Tristan Honscheid <honscheid@google.com>
The init function of the MAX17262 sensor doesn't check the return value
of its I2C read and write functions. In case a read operation fails, the
output variable is not updated but the driver proceeds anyways. This can
cause unintended operation due to the potentially un-initialized memory,
such as getting stuck in the polling loop on line max17262.c:245. Update
the init function to abort and pass along the error code when I2C
transactions fail.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
Fixes sensor drivers to consistently return -ENOTSUP when an unsupported
channel argument is passed to the sensor_channel_get function.
Signed-off-by: Maureen Helm <maureen.helm@analog.com>
The driver would return the temperature for all channels requested
except relative humidity. Instead, ENOTSUP should be returned for
unsupported channels.
Signed-off-by: Josep Puigdemont <josep.puigdemont@gmail.com>
This PR introduces a backend API to be implemented by sensor emulators
that creates a standardized mechanism for setting expected sensor
readings in tests. This unlocks the ability to create a generic sensor
test that can automatically set expected values in supported sensor
emulators and verify them through the existing sensor API. An
implementation of this API is provided for the AKM09918C magnetometer.
A generic sensor test is also created to exercise this implementation.
Observe that this test knows nothing about the AKM09918C; info about
supported channels and sample ranges is discovered through the backend
API. The test iterates over all devices attached to the virtual I2C and
SPI buses in the test binary's device tree, which (theoretically) covers
all sensors. Sensors whose emulator does not exist yet or does not
support the backend API are skipped.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
The sensor driver uses the value 500 to convert bit counts to microgauss
values, but it should actually be 1500. Edit the driver unit test to use
the macro instead of a magic number.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
Return ENOTSUP when getting the humidity channel if the driver is used
with a BMP280, since this device does not provide humidity readings.
Signed-off-by: Josep Puigdemont <josep.puigdemont@gmail.com>
Add a mutex to protect shared data-structures, since shell can have
multiple backends.
Signed-off-by: Marco Argiolas <marco.argiolas@ftpsolutions.com.au>
Added a driver for the DFRobot A01NYUB distance sensor. This sensor
sends its readings via UART at 9600 baud. This driver uses interrupts
to read the data from the sensor.
Signed-off-by: Oliver King <oliver.king@steadconnect.com>
Refactor the ESP32 target SOCs together with
all related boards. Most braking changes includes:
- changing the CONFIG_SOC_ESP32* to refer to
the actual soc line (esp32,esp32s2,esp32s3,esp32c3)
- replacing CONFIG_SOC with the CONFIG_SOC_SERIES
- creating CONFIG_SOC_FAMILY_ESP32 to embrace all
the ESP32 across all used architectures
- introducing CONFIG_SOC_PART_NUMBER_* to
provide a SOC model config
- introducing the 'common' folder to hide all
commonly used configs and files.
- updating west.yml to reflect previous changes in hal
Signed-off-by: Marek Matej <marek.matej@espressif.com>
- use CONFIG_HAS_HW_NRF_* symbols consistently in nRF multi-instance
drivers when creating particular driver instances
- remove unnecessary hidden Kconfig options that indicated the type of
peripheral to be used by a given instance (e.g. SPI, SPIM, or SPIS)
and enabled proper nrfx driver instance; instead, use one option per
peripheral type and include the corresponding shim driver flavor into
compilation basing on that option (not the one that enables the nrfx
driver as it was incorrectly done so far in some cases)
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
This fix a build issue where ICM42688_DECODER is enabled but
SENSOR_ASYNC_API is not, which results in some structures in an
orphan section:
warning: orphan section
`._sensor_decoder_api.static.invensense_icm42688__decoder_api_' from
`libdrivers__sensor__icm42688.a(icm42688_decoder.c.obj)'
being placed in section
`._sensor_decoder_api.static.invensense_icm42688__decoder_api_'
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Use a combination of sys_cpu_to_be16 and i2c_burst_write_dt for setting
16 bits registers. Get rid of a bunch of temporary variables, custom
conversions and few bus writes.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
The current ID for BQ274XX_DEVICE_ID is actually the one for the
BQ27421. The driver seems to work with the BQ27427 as well, at least the
common and extended commands are the same, so add that variant as well,
rename the existing one and print the currently read ID when the ID
check fails.
The configuration registers have a different offset though, so add a
register offset table and make the device rcognize the right one un
runtime based on the device ID.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Use i2c_reg_write_byte_dt instead of bq274xx_cmd_reg_write. The wrapper
does not add anything anyway.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Change bq274xx_sample_fetch to support SENSOR_CHAN_ALL. This makes it
possible to get the sensor data using the sensor shell, besides being
generally convenient.
Also drop a redundant comment.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
The device technical reference manual says "The Sealed to Unsealed key
has two identical words". Use two different defines with the same value
in the code so it's somewhat less ambiguous that the double write is
intentional.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Add a retry count limit to config update mode loops, this way the system
can still boot if there's an issue with the device. The normal sleep
should be enough for correct operation, adding a conservative limit.
Rework the delays to be unambiguous while at it.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Fix the calculation for designenergy_mwh, as right now it's using a
float casted straight to an int, which results in the factor rounded
from 3.7 to 3. Also rework both that and taperrate so that they don't
use floating point.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Control reg write seems to support two bytes mode (the technical
reference shows example of that), so use a single i2c_write_dt there.
Also drop a couple alias variables from bq274xx_cmd_reg_write.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Add the bq274xx_ back to the static function prefixes for the bq274xx
driver.
These have been removed recently but every other sensor and most Zephyr
driver have static function prefixed, this helps avoiding ambiguity in
list files, stack traces, setting debugging breakpoints etc.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
channel_pos_read bit was being cleared with a sensor channel enum value
instead of the encoded bit position.
Signed-off-by: Al Semjonovs <asemjonovs@google.com>
Forcing the usage of the RC oscillator and keeping it on turns out
to have detrimental effects to the readings by default. The default
clock mode settings are perfectly fine.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
When CONFIG_PM_DEVICE is enabled, the FDC2x1x driver includes code that
doesn't access the name of the shutdown pin's GPIO port correctly.
Correct this so the code derefences the right struct members.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
The FDC2X1X driver depends on newlib in Kconfig. This prevents the
driver from being built in a native_posix testing environment, which uses
an external libc from the host. Allow the driver to be built with an
external libc as well.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
This adds support for the bosch bmi085 and bmi088. This also includes
support for data sync mode.
Signed-off-by: Ryan McClelland <ryanmcclelland@meta.com>
Check and propagate two return values.
Don't need to check return of the part id request, but make
sure that the value is initialized before the comparison.
Fixes#58575
Coverity-CID: 316152
Signed-off-by: Thomas Stranger <thomas.stranger@outlook.com>
The return value is consciously not checked, because the operation
is expected to fail. And the real request is executed afterwards.
Fixes#58593
Coverity-CID: 316443
Signed-off-by: Thomas Stranger <thomas.stranger@outlook.com>
The underlying HAL driver may improperly forward an samplerdy event even
if it's disabled in the configuration. Ignore the event to prevent error
logs until the issue is fixed in HAL.
Signed-off-by: Marek Pieta <Marek.Pieta@nordicsemi.no>
