apollo3 driver added PM to UART for only apollo3
but defined PM function for both causing an error
in twister
Signed-off-by: Richard Wheatley <richard.wheatley@ambiq.com>
In RISCV "sepc" is actually the name of an instruction.
As this prevents building. We should remove the line.
Signed-off-by: Sven Ginka <s.ginka@sensry.de>
This commit excludes the `m2gl025_miv` platform from running the
`kernel.common.stack_sentinel` test. It experiences the same issue as
`hifive1`, which was described in issue #66070.
Signed-off-by: Filip Kokosinski <fkokosinski@antmicro.com>
Automatically handle device runtime PM for all flash API calls.
Asynchronously release the device after a small delay to minimise power
state transitions under multiple sequential API calls (e.g. NVS).
Signed-off-by: Jordan Yates <jordan@embeint.com>
Remove `SPI_NOR_IDLE_IN_DPD` to simplify the possible transition states
in the `spi_nor` driver. This option was originally added 5 years ago
when device runtime PM was in a much less mature state.
I do not believe having a separate power management implementation for
this one in-tree driver is in the interests of the project.
The behaviour of `SPI_NOR_IDLE_IN_DPD` leads to extremly suboptimal
behaviour in use cases where multiple small reads are performed
sequentially, see #69588.
Removal of this option does not break the behaviour of any existing
applications, only increases current consumption in idle by ~10uA until
device runtime PM is enabled.
Signed-off-by: Jordan Yates <jordan@embeint.com>
When change unaware client send ATT request it shall get Database
Out of Sync error. Reading GATT database hash is an exception here.
This was affecting GATT/SR/GAS/BV-05-C qualification test case.
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
Background of this modification is to make clock control
driver code provided by Renesas vendor to support for Renesas MCU
on Zephyr.
Signed-off-by: Quy Tran <quy.tran.pz@renesas.com>
A 16-bit value built using byte shifts and ORs from a given
couple of lsb and msb bytes will result to be the same on both
little-endian and big-endian architectures, e.g.
uint8_t lsb, msb;
int16_t val;
/* val is the same number on both le and be archs, but has
different layout in memory */
val = (msb << 8) | lsb;
All the xyz_raw_get() APIs of stmemsc sensor module build the sensor
data using the above method and DO NOT hence require (it actually leads
to wrong values on big-endian machines) to use any le/be swap routines,
such as sys_le16_to_cpu().
Fix#75758
Signed-off-by: Armando Visconti <armando.visconti@st.com>
Updated the counter_mcux_lptmr driver to support multiple
instances of the lptmr peripheral. Also added a new
binding property to identify if the user is using
counter-mode or pulse mode. since we were previously using the
prescaler value to check this which could be wrong
if used as a division value for getting the freq.
Added a property that allows the user to decide
what the counter value in lptmr should be divided by.
Cleaned up INIT macro for lptmr.
Signed-off-by: Emilio Benavente <emilio.benavente@nxp.com>
dts: arm: nxp: mimxrt1180_evk: add GPT1/2 instance into devicetree
Enable GPT1/2 clock
Add GPT1/GPT2 instances
Set GPT2 as a counter, the default frequency is 240000000
Signed-off-by: Lucien Zhao <lucien.zhao@nxp.com>
The reset function uses uint32_t as its 2nd parameter, which leads to the
following compiler warning.
zephyr/drivers/mipi_dbi/mipi_dbi_stm32_fmc.c:176:18: warning:
initialization of 'int (*)(const struct device *, k_timeout_t)' from
incompatible pointer type 'int (*)(const struct device *, uint32_t)'
{aka 'int (*)(const struct device *, unsigned int)'}
[-Wincompatible-pointer-types]
176 | .reset = mipi_dbi_stm32_fmc_reset,
| ^~~~~~~~~~~~~~~~~~~~~~~~
When you look at similar drivers,
/drivers/mipi_dbi/mipi_dbi_smartbond.c#L126
/drivers/mipi_dbi/mipi_dbi_nxp_lcdic.c#L561
you notice they use k_timeout_t.
So the whole fix is to correct the type.
Signed-off-by: Thorsten Spätling <thorsten.spaetling@vierling.de>
Round up buffer size used in UDC_BUF_POOL_DEFINE() to respect the
required buffer granularity. The issue was observed with UAC2 explicit
feedback data, but the problem applies to any UDC_BUF_POOL_DEFINE() use.
In order for every buffer returned by UDC_BUF_POOL_DEFINE() to be both
aligned and to have required granurality, it is required to allocate the
buffers for ROUND_UP(size, LCM(UDC_BUF_GRANULARITY, UDC_BUF_ALIGN)).
Because we do not have Least Common Multiple nor Greatest Common Divisor
compile time macros, assume that granularity is multiple of alignment.
Validate the assumption with a build time assert. When we get a target
where this assumption fails we would have to come up with a solution to
compute LCM and/or GCD at compile time.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Remove the early max packet size check on endpoint activation, because
there are valid use cases where the max packet size is not multiple of 4
and the stack won't perform multi-transaction transfers. Example use
case is UAC2 explicit feedback endpoint that has Max Packet Size equal
3 when device is operating at Full-Speed.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Fail TxFIFO write if the endpoint is not activated, because there is
essentially no fifo assigned. Writing to not activated endpoint is
possible, but it does tend to corrupt fifo number 0 which is used for
control transfers. USB stack should not really be attempting to write to
endpoints that have not been activated (ep_enable called).
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Store SOF frame number and use the information to set the Even/Odd
microframe bit when enabling isochronous endpoints. Reject received
isochronous data if number of packets received does not align with
received PID.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
RxFLvl interrupt should only handle the actual data movement from RxFIFO
to the buffer. OUT transfers are completed in XferCompl handler both in
DMA and Completer mode. This reduces code size by avoiding having
separate code paths based on operating mode.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Function dwc2_set_dedicated_fifo() calls dwc2_flush_tx_fifo() before the
diepctl register is written. The register will be only written after the
dwc2_set_dedicated_fifo() finishes. Solve the problem by passing fifo
number directly instead of endpoint address to dwc2_flush_tx_fifo().
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Add all register bit defines necessary for isochronous transfers. Clean
up the endpoint transfer size register defines clearly separating IN and
OUT registers because they do use different bit fields. Add alternate
bit names for bits that do have different meaning based on configured
endpoint transfer type.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
The clang-format check outputs a git diff with surrounding context. It
naively removed the first and last 3 lines, but this does not work if
there are less lines.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
As the .bin & .hex build output is optional
and it can be disabled by CONFIG_BUILD_OUTPUT_BIN/HEX,
add support for the mandatory .elf build output
to the pyocd runner flash command.
Signed-off-by: Andrej Butok <andrey.butok@nxp.com>
The terminal offset for subcommands' help text isn't
computed correctly, fix it.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
- Use `SHELL_CMD_ARG_REGISTER` for the main cmd to state the
required number of arguments, this helps to remove the
runtime check from the command, and also print the help
message to the terminal when the argument count is
unexpected.
- Some changes to the help text that hopefully makes the
mandatory and optional arguments more obvious to the user
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Reset delay was not being calculated correctly in lcdic driver, the
ticks field needs to be accessed directly within the timeout structure
to calculate the correct delay time
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
Exclude RW612 based boards from the WiFi testsuite, as these boards
require binary blobs be downloaded in order to build as expected
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
Adds unittests for the CAP Initiator start procedure, simply
verifying that the procedure works as well as
testing invalid parameters.
This also allows us to remove the invalid behavior checks
from the babblesim test implementation.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The driver in tree is for u-blox M8 devices, not M10. The M10 series
devices (from Protocol Version 23.01) use a different, non backwards
compatible interface for configuring the modem behaviour.
Of the two boards tested in the original PR, the "VMU RT1170" is
explicitly listed as having a u-blox NEO-M8N modem, while I have
been unable to find any information online about the "FMURT6" board.
Leaving the naming as-is will cause problems when M10 drivers are
contributed.
Signed-off-by: Jordan Yates <jordan@embeint.com>
disk_access functions were called from an interupt context, thus
crashing on mutexes (no sleep is allowed in interrupt context).
The usage of a spinlock is permitted. The functions guarded by
the spinlock are executed very quickly anyway, so the spinlock
is applicable.
Signed-off-by: Alexi Demers <alexi.demers@axceta.com>
Usb "write" method never copies the input data buffer, it uses the
buffer pointer directly. The fact that it was on the stack made
the buffer ephemeral, and could disappear before the USB transfer
actually occurred. The data transmitted was then random. This commit
converts the transfer buffer to a static buffer, so that it always
exists for the duration of the USB transfer.
Known side-effect: if "read capacity" command is read twice
simultaneously, the same buffer will be used twice. As the capacity of a
USB drive should not change between 2 calls, this side effect can be
ignored.
Signed-off-by: Alexi Demers <alexi.demers@axceta.com>
This provides memory mappings with the ability to be initialized in their
paged-out state and be paged in on demand. This is especially nice for
anonymous memory mappings as they no longer have to allocate all memory
at mem_map time. This also allows for file mappings to be implemented by
simply providing backing store location tokens.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
