Fix an issue where the BME280 sometimes returns an incorrect chip ID
immediately after a power cycle. This causes sensor initialization to fail.
According to the datasheet, the sensor requires a 2 ms start-up delay after
power is applied. This patch introduces a sleep delay to ensure the
required start-up time is respected before reading the chip ID.
Signed-off-by: Filip Stojanovic <filipembedded@gmail.com>
Coverity reports a potential integer overflow in the accel_range
computation due to the use of a left shift on an int type.
CID 520269: Unintentional integer overflow (OVERFLOW_BEFORE_WIDEN)
Even though the register value is constrained to 0–3 by the BMA456 spec,
and no real overflow occurs, an explicit cast to int64_t prevents false
positives and aligns with safe coding practices.
Fixes: #90517
Signed-off-by: Gaetan Perrot <gaetan.perrot@spacecubics.com>
Add the 'const' qualifier when casting 'dev->config' to 'struct
bmp581_config *' to preserve const-correctness and fix SonarQube warning :
"A cast shall not remove any const or volatile qualification from the
type of a pointer or reference."
This improves maintainability by respecting the const contract and resolve
static analysis issues.
Signed-off-by: Gaetan Perrot <gaetan.perrot@spacecubics.com>
Move null check for 'dev' before dereferencing it to access dev->config.
This ensures the check is meaningful and avoids undefined behavior in case
of a null device pointer.
Signed-off-by: Gaetan Perrot <gaetan.perrot@spacecubics.com>
Add an explicit log message if setting suspend mode fails during error
handling in bmm350_init_chip(), to improve debuggability.
This addresses Coverity issue CID 520279 (Incorrect expression - CWE-398),
which flagged a conditional block where both branches effectively led to
the same outcome (return -EIO), making it appear redundant.
By adding a log before returning, we clarify the purpose of the condition
and avoid the issue of "identical code for different branches", while
keeping the functional behavior unchanged.
Coverity-CID: 520279
Signed-off-by: Gaetan Perrot <gaetan.perrot@spacecubics.com>
There were a lot of places where the return value was sum'ed up
which can be slightly easier to read, but can give an odd return
value in the end. This changes it to just return immediately if
an error is seen.
There was also runtime NULL checking of input arguments that are
rather programmer clown prevention. Change these to asserts.
The return value didn't have a consistant type through the file,
make it consistently an int.
This changes the name of the return value of fix_sign to be
signed_value differenating it from the ret name used elsewhere.
Use the sizeof(buffer) as the argument for reg reads and writes
rather than a hard number.
Some log messages ended with '\n'. Remove all these.
Signed-off-by: Ryan McClelland <ryanmcclelland@meta.com>
Add configurations for setting the pad drive strength and the
interrupt active high/low and od/pp.
Signed-off-by: Ryan McClelland <ryanmcclelland@meta.com>
There is a race condition where an interrupt can fire before the
drdy_handler is registered. The drdy_handler will tradionaly
callback the sample get clearing the interrupt... but if it's not
configured and NULL, the interrupt will stay forever latched. Read
the interrupt status to clear the interrupt flag allowing it to
trigger again.
Also, move the serial api function helpers in to the header
allowing them to be used from other c files.
Signed-off-by: Ryan McClelland <ryanmcclelland@meta.com>
Fix the oversampling value to be 1 from 0 for no oversampling to be
consistant with other sensor drivers.
Signed-off-by: Ryan McClelland <ryanmcclelland@meta.com>
Add the attr_get api. Also rename functions where they implied
accelerometer to imply magnetometer.
Signed-off-by: Ryan McClelland <ryanmcclelland@meta.com>
A write to the "ctrl_meas" register can cause the sensor to transition
from sleep to normal mode (which is default Kconfig settings). As per
BME280 datasheet, writes to the "config" register in normal mode may be
ignored. This can lead to use BME280_STANDBY_05MS and BME280_FILTER_OFF
instead of the values set by the application. To fix this, write to the
"config" register before "ctrl_meas" in the sensor init function.
Signed-off-by: Loic Domaigne <tech@domaigne.com>
Fix undeclared variable error by changing 'config' to 'cfg' in the
data_sync condition check, matching the variable name declared at the
beginning of the function.
Signed-off-by: Liang Jiaxiang <moonbite233@gmail.com>
Calling sensor API functions on devices not in `PM_DEVICE_STATE_ACTIVE`
is a violation of the PM API. Adding manual checks inside of drivers
complicates the drivers and increases ROM footprint for no additional
benefit.
Signed-off-by: Jordan Yates <jordan@embeint.com>
The bmp581 driver currently returns pressure in pascals, when
the sensor_channel enum specifies kilopascals.
Signed-off-by: Maxmillion McLaughlin <max@sorcerer.earth>
Since #83575 bma4xx driver relies on RTIO to work. This means that
CONFIG_I2C_RTIO should be defined together with CONFIG_I2C in the
Kconfig, otherwise I2C_DT_IODEV_DEFINE() won't be available at
build time.
Signed-off-by: Valerio Setti <vsetti@baylibre.com>
Clamp the returned humidity to a non-negative value. This ensure that
the return value is always within the expected range, even if the
calibration parameters are read out incorrectly or corrupted.
This check is applied in the example Bosch driver for the float
conversion case, so there are situations that warrant it.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Add a streaming implementation for the BMA4XX using both
FIFO watermark and FIFO full interrupts. A batch duration of
3,000 have been used for verification.
Signed-off-by: Peggy Chen <peggy.chen@cienet.com>
Updated the sensor configuration to use the bma4xx_safely_configure
function for safer and more reliable initialization.
Signed-off-by: Peggy Chen <peggy.chen@cienet.com>
Added register definitions, bit masks, and related constants
for the BMA4XX sensor in a separate header file.
Signed-off-by: Peggy Chen <peggy.chen@cienet.com>
1. Temperature Interface
According to BMI08x datasheet, temperature reading
requires both MSB and LSB bytes to be read and
processed correctly.
Temp data processing should follow the formula:
Temp in °C = (temp_msb * 8) + (temp_lsb / 32)
This patch implements the correct reading
sequence and calculation method as specified
in the datasheet.
2. Trigger Setting
Previously we set handler and then trigger struct.
However under some situation, as long as we set
the handler, we get into ISR immediately and never
set trigger struct.
I simply changed the sequence.
Testing:
- Verified temperature readings match datasheet
- Tested on stm32f407igh board with BMI08x sensor
Fixes: #82375
Signed-off-by: Wenxi Xu <xuwenxi0517@gmail.com>
This commit introduces a new Sensor Clock API, enabling the retrieval
of cycle counts and conversion to nanoseconds based on the system or
external clock. The API includes:
- `sensor_clock_get_cycles()` to get the current cycle count from the
sensor clock.
- `sensor_clock_cycles_to_ns()` to convert cycles to nanoseconds using
the clock's frequency.
The implementation supports both system clocks and external clocks
defined in the device tree, making the sensor clock integration more
flexible for various sensor use cases.
Signed-off-by: Mark Chen <mark.chen@cienet.com>
Wait until the typical time that the sample would be ready before
starting to poll the status register.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Add a timeout to the `bme280_wait_until_ready` loop to ensure that we
never get stuck in this loop even if the bus enters an error condition.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Add support for Bosch bmp390 sensor, the drop in replacement for the
bmp388 with same register but different chip-id. This patch make use
of "bosch_bmp390" or "bosch_bmp388" and set the specific chip-id in a
data->chip-id variable, which then used to check against the register
value.
Additional, manual shift operation had been replaced with ENDIAN safe
macros and calibration values with target variable of int16_t add a
cast for it.
bmp388_spi: read register implementation wrong, fixed it.
tx-buffer must be <addr><dummy><dummy> in order to receive the
register value. Read registers in burst mode and have rx and tx
buffer same spi_buf to avoid clock stop and delay with nrf5.
Signed-off-by: Chris Ruehl <chris@gtsys.com.hk>
Inform the executor of a submissions completion with -ENOMEM
if the size of the workq is not big enough.
Signed-off-by: Florian Weber <Florian.Weber@live.de>
Support the BMM150 being on a power domain, which may not be powered at
boot. For example, Nordic Thingy53.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Update power management to only start/stop periodic measurements in
`CONFIG_BME280_MODE_NORMAL`, instead of re-initialising the chip
completely.
In `CONFIG_BME280_MODE_FORCED`, there is nothing to do when suspending,
as the sensor is already in its lowest power mode.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Add missing braces to comply with MISRA C:2012 Rule 15.6 and
also following Zephyr's style guideline.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
