This commit finish to fix the bug describe by 85e2a0679a68f02f7ef.
With the previous correction, the uptime read could be in the past:
if the counter rewinds just after testing ARRM flag, we had lost
some counts.
Signed-off-by: Julien D'Ascenzio <julien.dascenzio@paratronic.fr>
Move to using 'select SPI' instead of 'depends on SPI'
(see commit df81fef for more details)
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
The interrupt status of the GPIO was not cleared when a new
interrupt configuration was set. This prevents the driver from
passsing all the gpio tests.
Fixes#54833
Signed-off-by: Lucas Tamborrino <lucas.tamborrino@espressif.com>
Note: RP2040 can support single-ended IO, by setting the GPIO_OUT
register to a constant value, and then changing the GPIO_OE register
instead, when the output has to change. To do this, the output-setting
functions need to know which pins have been configured as single-ended,
and for that reason the data structure has been extended to include
this information.
Another change is the PR, is that setting of the pull-ups/pull-downs
now applies to both inputs and outputs as well. Previous solution was
wrong, because if the user wanted to set up an input with a pull
resistor enabled, and then reconfigure it to an output without any
pulls, then the pulls remained in place for the output. Now pulls
are correctly set based on the gpio flags for outputs too, and this
is especially useful for single-ended outputs too.
Signed-off-by: Purdea Andrei <andrei@purdea.ro>
Populate the `channel` index when constructing configuration structs for
secondary RTT channels. Originally missed in #27704.
Fixes#54955.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Add the implementation for the akm09918c magnetometer driver.
Additionally, add the appropriate node to the TDK robokit1 device
tree. In order to prevent regressions, add the sensor to the sensor
build_all tests and specific tests using an emulator.
Signed-off-by: Yuval Peress <peress@google.com>
Add a driver implementation that uses the I2S peripheral.
Based off this blog post:
https://electronut.in/nrf52-i2s-ws2812/
Should help with #33505, #29877 and maybe #47780, as there is no garbage
data at the end of transmissions on nRF52832, and no gaps.
Signed-off-by: Jonathan Rico <jonathan@rico.live>
The icm42688 introduced a safe configure function to make the
configuration more atomic, but the main driver was never updated
to call it so the new config values were discarded.
Signed-off-by: Yuval Peress <peress@google.com>
Avoid implementations in the subsys emul directory to keep the directory
focused on emulator subsystem instead of consumers. Consumers should be
implemented side-by-side to their drivers
Signed-off-by: Yuval Peress <peress@google.com>
Allow the emulator to sit in the same directory as the driver
implementation. This will make working on the emulator much easier and
keep the emulator subsystem directory clean by allowing it to focus on
the actual subsystem and not the use cases of the subsystem.
Signed-off-by: Yuval Peress <peress@google.com>
No need to read back the CR (NOR flash config or control register)
to check if entering the 4-Byte Address Mode is effective.
The action of this command is immediate and the result
(bit field of the CR) is NOR flash vendor-specific.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
Drop STM32 pinmux driver in favor of pinctrl. Some definitions located
in pinmux headers were used by the pinctrl driver, so they have been
moved there.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Use proper format for size_t type eliminating warnings of type:
...
error: format '%u' expects argument of type 'unsigned int', but
argument 2 has type 'size_t' {aka 'long unsigned int'}
[-Werror=format=]
...
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Commit 00adb2a539 ("drivers: udc: remove no more required pending
state flag") removed state parameter from udc_buf_peek and udc_buf_get.
Adapt udc virtual driver so the code compiles.
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
STM32L4plus mcu has SDMMC internal DMA which works without any
configuration and it's handled by SDMMC HAL driver. This commit adds
option to enable it and use it.
Signed-off-by: Petr Hlineny <development@hlineny.cz>
The LoRa modem has been staying in STATE_BUSY
after sx12xx_lora_test_cw() call
because Radio.SetTxContinuousWave() doesn't configure the PHY
to raise 'TX done' event.
It runs TxTimeoutTimer instead,
so we have to handle 'TX timeout' event to correctly release the modem.
This fix has been tested on custom board with STM32L4 and SX1276.
Signed-off-by: Petr Sharshavin <sharshavin@mail.ru>
Move the can_sja1000.h header file to the public
include/zephyr/drivers/can/ include path. This allows writing out-of-tree
SJA1000 based driver front-ends.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Rename the can_utils_filter_match() utility function to
can_frame_matches_filter(), move it to the public
include/zephyr/drivers/can.h header file, and add a missing mismatch for
standard (11-bit) ID frames passing extended (29-bit) filters.
This allows using the utility function in out-of-tree CAN drivers.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
When checking if any UART TX/RX IRQs are pending, the driver should also
consider whether these IRQs are enabled. Or we still get pending status
set even if the related interrupts are disabled.
Signed-off-by: Mulin Chao <mlchao@nuvoton.com>
In order to avoid concurrency situation during writing UFTCTL register,
this CL adds critical sections to prevent the unexpected result if the
driver wants to set/clear bits of this register.
Signed-off-by: Mulin Chao <mlchao@nuvoton.com>
Signed-off-by: Jun Lin <CHLin56@nuvoton.com>
It is advised to stop any ongoing ADC conversion before
disabling the peripheral.
Added some comments so that ADC state isn't left into
intermediate state that would prevent correctly
enabling or disabling the peripheral
Signed-off-by: Cyril Fougeray <cyril.fougeray@worldcoin.org>
This is necessary to prevent stack overflows when building with
non-standard configurations (e.g. CONFIG_DEBUG).
Adding them as hidden kconfig options to avoid the stuck kconfig syndrome.
Users are free to redefine them in their app to force a value.
The userchan.c driver is only built for posix, and the help text for
ARCH_POSIX_RECOMMENDED_STACK_SIZE states that the real stack comes from
somewhere else (the pthread stack), hence why it doesn't use the new
kconfig options.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
There are just list of child files to include, right now there's a hint
of that trying to be alphabetical but then entries have been added with
various logic, so one has to figure where stuff has to be added.
Rewrite these to:
- keep the list alphabetical, for CMakeLists unconditional first,
conditional after so one just has to select and sort
- drop the spaces in the Kconfig
- use a single space between target and directory
- rename from "Sub Systems" to "Subsystems" (single word)
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
The fuel gauge API doesn't make use of syscalls with device drivers, which
is not congruent with Zephyr's security model.
Correct the fuel gauge API to only allow access to fuel gauge device
drivers functionality via syscalls to keep with Zephyr's security model.
Signed-off-by: Aaron Massey <aaronmassey@google.com>
Fuel gauge API struct type is currently named battery_driver_api, this is
inaccurate as the API struct represents functions for interacting with fuel
gauge drivers in particular.
Rename the API struct type to fuel_gauge_driver_api to clearly signal its
functionality.
Signed-off-by: Aaron Massey <aaronmassey@google.com>
The int config and fifo config register addresses were wrong, fix those.
Adds lots of debug information (when LOG_DBG=y) to the configuration of the
device which is incredibly helpful for diagnosing configuration issues.
Disables the device interrupts while reconfiguring. Adds a safely
reconfigure function which will rollback to previous configuration
on misconfiguration.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Several issues:
- `last_count` should not be updated with current time or this will
cause a time drift and induce jitter due to IRQ servicing latency.
- `sys_clock_set_timeout()` should not base its `mtime` on the current
time either. Tracking the `last_tick` and `last_elapsed` values avoids
the need for all the tick rounding computation.
- The MIN_DELAY thing is pointless. If the delay gets close or even behind
current time then the IRQ will be triggered right away. This is unlikely
to happen very often anyway so the constant overhead is uncalled for.
- Runtime 64-bits divisions on 32-bits hardware are very expensive.
Fix the above, and improve the following:
- Prime the accounting by simply invoking the IRQ handler from the init
code. That will make the "ticks since boot" counter right.
- Remove excessive casts, especially a few wrong ones.
- Simplify the code overall.
Here's the output from the timer_jitter_drift test.
Before this patch:
|timer clock rate 60000000, kernel tick rate 10000
|period duration statistics for 10000 samples (0 rollovers):
| expected: 1000 us, 60000.000000 cycles
| min: 907.600000 us, 54456 cycles
| max: 1099.750000 us, 65985 cycles
| mean: 1008.594633 us, 60515.678000 cycles
| variance: 2.184205 us, 7863.136316 cycles
| stddev: 1.477906 us, 88.674332 cycles
|timer start cycle 995589, end cycle 606152369,
|total time 10085946.333333 us, expected time 10000000.000000 us,
|expected time drift 0.000000 us, difference 85946.333333 us
After this patch:
|timer clock rate 60000000, kernel tick rate 10000
|period duration statistics for 10000 samples (0 rollovers):
| expected: 1000 us, 60000.000000 cycles
| min: 992.116667 us, 59527 cycles
| max: 1030.366667 us, 61822 cycles
| mean: 1000.001902 us, 60000.114100 cycles
| variance: 0.105334 us, 379.201081 cycles
| stddev: 0.324551 us, 19.473087 cycles
|timer start cycle 987431, end cycle 600988572,
|total time 10000019.016667 us, expected time 10000000.000000 us,
|expected time drift 0.000000 us, difference 19.016667 us
The mean, variance and standard deviation number differences speak for
themselves, even in the absence of competing ISRs and/or IRQ-disabled
periods which would have made the comparison even worse.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
