clock control is required for "fast instances" so assert clock
is enabled alongside PM DEVICE RUNTIME. Update UART tests to
reflect this requirenment.
Signed-off-by: Bjarki Arge Andreasen <bjarki.andreasen@nordicsemi.no>
Currently there is a mismatch between the naming of the hardware and
the drivers targetting the hardware. nrf2_ is used instead of
the actual bindings names, like nrf2_audiopll instead of
nrfs_audiopll. This makes it hard to map drivers to the hardware
they are targetting.
There is historical reason for some of this, namely the same binding
name was used for different hardware, which is why nrf2_ was used
on newer platforms. This is no longer the case though, so drivers
and configs can be named according to the hardware without conflict.
Signed-off-by: Bjarki Arge Andreasen <bjarki.andreasen@nordicsemi.no>
When reception is restarted (STARTRX after ENDRX but no STOPRX) it is
possible that FRAMETIMEOUT countdown counter will not be started by
the first received byte if byte was already being transmitted when
STARTRX was called. If that is the only byte then it is expected that
timeout will be triggered but since FRAMETIMEOUT counter is not started
there is no FRAMETIMEOUT event which has short to STOPRX. This
situation will happen in case short buffers are used (< 5 bytes)
because then short ENDRX_STARTRX is not used then.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Add support for getting to the lowest power mode when polling is
used with disable-rx property and interrupts are not used for
that UARTE. So far disabling of the UARTE peripheral was done in
the interrupt but in some cases interrupt may not be available
and in that case uart_poll_out shall wait until byte is transferred
and put UARTE into the lowest power state.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Some SoCs generates unexpected RXTO event during restart.
Restart happens when ENDRX_STARTRX short is enabled and STOPRX
is triggered (via short or by CPU). STOPRX starts closing
procedure and ENDRX event is generated at some point which
triggers STARTRX and closing procedure is interrupted. RXTO
should not be triggered in that case. Due to internal timings
some SoC on fast UARTE instance will trigger RXTO followed
by RXSTARTED. This RXTO event shall be cleared as receiver is
actually restarted and not stopped.
Affected SoC is not in tree so Kconfig is added which enables
the workaround.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
PM_DEVICE_ISR_SAFE shall not be used when non-asynchronous API is used
because RX is disabled in suspend action and that takes relatively
long time. In case of PM_DEVICE_ISR_SAFE it is done with interrupts
disabled. RX is not used at all if disable-rx property is set and in
that case PM_DEVICE_ISR_SAFE can be used.
Added macro which determines if ISR safe mode can be used.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Setting frame size and endianess is now supported in UARTE HAL,
so these settings must have default values configured
to avoid compiler warnings about uninitialized fields.
Signed-off-by: Nikodem Kastelik <nikodem.kastelik@nordicsemi.no>
Fast instance in nrf54h (uart120) can generate a spurious RXTO event
some time after RXTO event that indicates that RX path is disabled.
The time when event is generated depends on baudrate and when slower
baudrates are used peripheral is disabled on time to not notice it
in the test but with higher baudates issue become visible. In order
to avoid spurious interrupt, RXTO interrupt is disabled during RXTO
event handling and enabled when RX is enabled. This workaround is
applied only for fast instance to avoid unnecessary register
accesses for slower instances.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
nrf54x devices have UARTE instance capable of using baudrate higher
than 1M. Higher baudrates does not have predefined values for
BAUDRATE register. A formula can be used to calculate BAUDRATE
value that shall be used for desired baudrate. Add UARTE_ANY_HIGH_SPEED
macro which is set when high speed is enabled (uart00 in nrf54lx or
uart120 in nrf54h20). For high speed instance use formula for getting
value that shall be written to BAUDRATE register.
When runtime configuration is not used then same formula is used to
calculate fixed BAUDRATE value.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Rename UARTE_ANY_FAST to UARTE_ANY_FAST_PD. There are 2 types of
'fast' UARTE instances. In nrf54h20 instance uart120 is in fast
power domain that requires additional power and clock management
of that domain. In nrf54lx fast uart00 instance does not require
that. Add _PD to indicate fast power domain.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Request fast global domain to run at 320 MHz during fast UARTE
activity. As request is asynchronous it cannot be called from
an ISR. Due to complexity to handle that without device runtime
power management a requirement is added so that if fast UARTE
is used device runtime PM must be enabled. Clock is request
and released in PM resume and suspended actions which in case
of fast UARTE are only called from thread context.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Low power mode for non-asynchronous API case is only available when
RX is not used (RX pin is not defined). In that case TX starting
function was using uarte_enable_locked() which tracks if UARTE is
used by TX or RX and TXSTOPPED interrupt was disabling UARTE
unconditionally. Because of that following attempt to TX start
was assuming that UARTE is already enabled when it was disabled.
Fixing it by using uarte_disable_locked function in TXSTOPPED
handling.
Code reordering was required to make uarte_disable_locked()
available earlier.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
The assert BUILD_ASSERT(NRF_GPD_FAST_ACTIVE1 == 0); is not correct
given that NRF_GPD_FAST_ACTIVE1 is defined as 1U, and is not used
in the file anyway. Remove the build assert.
Signed-off-by: Bjarki Arge Andreasen <bjarki.andreasen@nordicsemi.no>
When fast UARTE instance is used (e.g. UARTE120 in nrf54h20), PM actions
are not ISR safe because they include communication over IPC so they can
only be called from the thread context. Extend driver to support both
PM modes. When non ISR mode is used then uart_rx_enable() and uart_tx()
will return error if they are called from ISR and resume operation
would need to be called because device is suspended. On completion,
driver is calling pm_device_runtime_put_async which can be called from
the ISR context.
Additionally, suspending in the TXSTOPPED and RXTO events has been
moved after user callback. It allows to support the case where
uart_rx_enable() or uart_tx() are called from that callback context.
Since suspending is called after returning from the callback it will
not trigger suspend action because API called in the callback context
will increment the usage counter (when pm_device_runtime_get() is
called).
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Or more precisely, do not remove the workaround for them.
The current HW models are accurate enough in this area for the
workaround to work properly.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
TX part of interrupt driven API was not calling PM device runtime
API. Additionally, when txstopped occurred after poll out it was
not handled correctly.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
b6d45423c6 Added support for pin retention but it was added
only to the case when device PM is used. There is another mode
in which UARTE is disabled when idle (low power mode) and in
that case pin retention must also be added.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
RX FIFO flushing on RXTO event should only be performed when
UARTE peripheral might be disable during inactivity and that
happens when low power modes is enabled or when device runtime
PM is used. Flushing was incrementing flush_cnt which was not
used (flushed data is not copied to the next buffer) which
was causing data loss and invalid RX data length reporting.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
When GPD is managed by pinctrl, pins retention needs to be controlled by
the driver to avoid glitches.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Optimize function which enables UARTE peripheral. It is called
only when interrupts are locked so data->flags does not require
atomic operation. Use standard logical operations so save few
bytes.
Simplify uarte_disable_locked.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Improved a workaround. When FIFO is not empty then FLUSHRX
task will generate RXSTARTED event to indicated that DMA transfer
was started. This property can be used to detect when FIFO was
not empty and workaroud a HW bug where RX.AMOUNT register is not
updated after flushing empty FIFO.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Add runtime PM to the driver. When asynchronous or interrupt
driven API is used, there are 3 independent paths for enabling
the device: RX, TX and TX poll_out. TX poll_out requires special
handling because number of poll_out calls does not need to much
number of TXSTOPPED interrupts. To handle that special flag is
added. For standard RX and TX is simpler.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
A code ifdefed by UARTE_BAUDRATE_RETENTION_WORKAROUND was missing a
closing bracket. It was missed because CI did not compile any target
that is applying this workaround and enables code compilation.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Add lock to fix race when uart_rx_disable is interrupted by RXTO
event which lead to driver state corruption.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
On uart120 BAUDRATE register is not retained when ENABLE=0 and
because of that BAUDRATE must be set after enabling. Add workaround
to the driver.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Add configurable magic byte instead of fixed 0xAA. In some cases, it
is known that certain bytes are less likely in the transmission and
picking specific magic byte may reduce probability of failure of
detection of the correct amount of flushed data.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Add support for DMM which manages cache and dedicated memory spaces.
Added support for data cache for buffers which are not DMM managed.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Frame timeout is a hardware feature present in newer versions
of UARTE (e.g. in NRF54X platforms) for detecting idle state
on RX line and ending RX after configurable timeout.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Rework driver to support new way of asynchronous RX handling.
Previously RX was handled in two modes: using RXDRDY interrupt for byte
counting or TIMER + PPI. Both modes had flaws. RXDRDY interrupt mode
could miscalculated amount of received bytes when interrupt was not
handled on time. Data was not lost but was not reported on time that
could lead to issues. PPI+TIMER mode requires additional resources
thus it was not the default mode. Often user was not aware of that
option and was expiriencing driver RX faults.
New RX mode is switching buffers when there is new data (RXDRDY event
not set for given amount of time). It does not require additional
resources to get precise byte counting. Additionally, this is in line
with new UARTE feature (RX frame timeout) which is present in nRF54X
devices. The behavior of the driver is the same for legacy devices
and new one. For legacy devices k_timer periodic interrupts are used
to check if there are any new bytes and it is not needed when RX frame
timeout is present.
Improved RX mode is enabled by default
(CONFIG_UART_NRFX_UARTE_ENHANCED_RX=y) but legacy modes are still
available though not recommended to be used.
Note that new RX mode (CONFIG_UART_NRFX_UARTE_ENHANCED_RX=y) behaves
a bit different because timeout always triggers switch of buffers
which means that there will be no UART_RX_RDY events with non-zero
offset. It also means that every UART_RX_RDY will be followed by
UART_RX_BUF_RELEASED.
After rework, driver is recommended to be used for all platforms as it
performs much better and takes much less code than the second UART shim
available for Nordic devices.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Refactor RX asynchronous API function to use a pointer to the RX
async data structure instead of top level data structure pointer.
It improves readability with more concise code.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Update nrfx drivers with the minimum number of bytes that can be sent
in a single call to `uart_fifo_fill`.
Signed-off-by: Jordan Yates <jordan@embeint.com>
This is a leftover from pre-pinctrl era and no longer makes sense.
Driver always manages gpio through pinctrl.
Support removed from uart and uarte shims.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
When rx_flush() was called with NULL buffer it was an indication to
drop data from the FIFO. However, it is still important to get the
correct amount of dropped data because when PPI+TIMER are used to
count bytes those flushed bytes are also counted (because each
byte generates RXDRDY event). If those bytes are not counted then
total amount of reported bytes is not correctly aligned with bytes
counted by TIMER.
Reworking rx_flush() to correctly count number of dropped bytes and
then add this flushed bytes to the total amount of RX bytes
reported to the user in RXTO event handler.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
When interrupt driven instance is using only TX then low power mode
is automatically enabled and then TXSTOPPED interrupt is disabled in
idle so uarte_nrfx_irq_tx_ready_complete() was returning wrong value.
Adding a dedicate flag which holds information is user enabled TX
interrupt.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
The initialization of the UARTE is attempting to handle potential
improper handover from the bootloader. This handling is additional
complexity which should be fixed in the bootloader or whatever
component is failing to deinit the UART properly.
This commit removes the handling of improper handover from the
bootloader, while additionally clearing the ERROR flag when suspending
the UARTE component.
Signed-off-by: Bjarki Arge Andreasen <bjarki.andreasen@nordicsemi.no>
Code size can significantly reduced (220 bytes per instance) if data
structure (stored in RAM) is not initilized (can be moved to .bss).
Data for asynchronous API had two fields which can easily be moved
to the configuration structure (which is in ROM) because they do not
change duing runtime.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
When runtime configuration is not enable then we can determine at
compile time what are the hardware settings. Function which
translates zephyr values to nRF register values is not needed as
macros can do that at compile time.
This optimization saves almost 400 bytes of code.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Some UARTE instances may be clocked at higher speeds than 16MHz, so the
baudrate setting needs to be scaled accordingly. This patch parses the
`clocks` property and obtains the clock-frequency property of the
associated clock, assuming it is a fixed-clock. We should ideally have a
proper clock control subsystem where frequency can be queried using an
API, but we're far from there.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
This is needed to avoid warnings about uninitialized
structure member, which was added in nrfx 3.6.
Signed-off-by: Nikodem Kastelik <nikodem.kastelik@nordicsemi.no>
When low power mode is enabled then whenever UARTE is not active,
driver attempts to put the peripheral into the lowest power state
by stopping and disabling UARTE. However, it did not put pins into
sleep state which could lead to increased current consumption.
Adding pins state handling to the low power mode.
Pins are put into sleep state only if CONFIG_PM_DEVICE=y.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Legacy shim takes less flash so it should be a first choice on cores
with less code memory (like RISCV cores on nrf54h20). Adding new
instances support to the legacy shim.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
Bjarki Arge Andreasen
Krzysztof Chruściński
Nikodem Kastelik
Pieter De Gendt
Alberto Escolar Piedras
Gerard Marull-Paretas
Jordan Yates
Anas Nashif
Hess Nathan