Add a new boolean devicetree property `tx-rx-swap` to the
st,stm32-usart binding, used to control TX/RX swap during
device initialization.
Signed-off-by: Alexander Mihajlovic <a@abxy.se>
The pm_constraint_* APIs were effectively used by the policy manager
only. This patch renames the API to the policy namespace and makes its
naming more explicit:
- pm_constraint_set -> pm_policy_state_lock_get()
- pm_constraint_release -> pm_policy_state_lock_put()
- pm_constraint_get -> pm_policy_state_lock_is_active()
The reason for these changes is that constraints can be of many types:
allow/disallow states, impose latency requirements, etc. The new naming
also makes explicit that the API calls will influence the PM policy
behavior.
All drivers and documentation have been updated accordingly.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
When CONFIG_PM=y, the IRQ function also needs to be stored, something
the "generic" uart_device_config cannot support. This uses the config
irq_config_func field to store the function in all situations, thus
removing unnecessary logic and finally dropping uart_device_config.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Instead of using "generic" uart_device_config fields, store the right
pointer to avoid unnecessary casts. This change makes code simpler and
more idiomatic.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
A common pattern here was to take the work item as the subfield of a
containing object. But the contained field is not a k_work, it's a
k_work_delayable.
Things were working only because the work field was first, so the
pointers had the same value. Do things right and fix things to
produce correct code if/when that field ever moves within delayable.
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Gives more explanation on uart_stm32_err_check function.
On stm32 F4X, F1X, and F2X, when clearing the usart Error Flag
(PE, ORE, FE, NE), the LL Clear function applies a software sequence
which reads the usart SR then the usart DR.
Consequently the RXNE flag is affected (cleared) by the
uart_stm32_err_check function call.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
The current driver doesn't handle the LBD flag, this leads
uart_stm32_err_check to return always true if a Line Break
is detected.
This PR adds Line Break Detection and the related flag clearing,
F0 series it's excluded from the changes.
Fixes zephyrproject-rtos#41339
Signed-off-by: Andrea Campanella <andrea.campanella@helvar.com>
Stop using redundant DEV_DATA/DEV_CFG macros and use dev->data and
dev->config instead.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
An additional devicetree poperty `single-wire` is added
to uart and usart bindings of stm32. The driver checks this value
during initialization and enables the single wire mode when set.
Signed-off-by: Jonathan Hahn <Jonathan.Hahn@t-online.de>
The PINCTRL_DT_(INST_)DEFINE macros already defined the trailing ;,
making its usage inconsistent with other macros such as
DEVICE_DT_DEFINE.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Some tests like:
tests/kernel/sched/metairq/kernel.scheduler.metairq
tests/kernel/profiling/profiling_api/kernel.common.profiling
tests/kernel/sched/schedule_api/kernel.scheduler
tests/kernel/sched/schedule_api/kernel.scheduler.multiq
tests/kernel/profiling/profiling_api/kernel.common.profiling
tests/kernel/workq/work_queue/kernel.workqueue
don't support that the current thread change when writing a message with
printk (which uses poll_out). So, we remove the call to k_yield which is
useful only for optimizing cpu usage by forcing a thread change if the
usart send stack is full.
Signed-off-by: Julien D'ascenzio <julien.dascenzio@paratronic.fr>
A dead lock could happen if 2 threads with differents priorities use
poll_out. In fact, the lock data->tx_lock could be lock by a thread with
lower priority and then a thread with higher priority can't take the
lock. There was a race condition here:
/* Wait for TXE flag to be raised */
while (1) {
if (atomic_cas(&data->tx_lock, 0, 1)) {
/* !!!!!!!! RACE CONDITION !!!!!!!!!!!!!!
if (LL_USART_IsActiveFlag_TXE(UartInstance)) {
break;
}
atomic_set(&data->tx_lock, 0);
}
}
To fix race condition, the interrupts are locked in poll_out.
Signed-off-by: Julien D'ascenzio <julien.dascenzio@paratronic.fr>
A lock was added to manage situation where the API poll_out and irq API
are used in same time.
Signed-off-by: Julien D'ascenzio <julien.dascenzio@paratronic.fr>
Current driver set a fixed prescaler value for the lpuart
that caused certain baudrate configurations to fail due to
LPUARTDIV overflow the LPUART_BRR register.
This PR attempt to calculate a suitable PRESCALER for the
selected baudrate, throws error and return if it couldn't get
an optimal one.
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
For some stm32 soc devices, the USART (or UART) flag RXNE is cleared
by the LL_USART_ClearFlag_RXNE function which directly writes
the RXNE bit of the Status register. This is the case with the
stm32F1x, stm32F2x,stm32F4x, stm32L1x.
Some other are using the Rx Data Flushing function.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
When the "Read data register not empty" irq occurs,
this commit is cleaning the RXNE flag by flushing the RX register
since the Receive Data Reg. (USART_RDR) has not be read previously
This could be the case when aborting a Rx for example.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
If a transmission is made with poll_out and immediately after an other
transmission is made with interrupt api the transmission is locked.
We fix this behavior by clearing the tx_poll_stream_on flag during the
irq_tx_enable function
Signed-off-by: Julien D'ascenzio <julien.dascenzio@paratronic.fr>
Refactors all of the serial drivers to use a shared driver class
initialization priority configuration, CONFIG_SERIAL_INIT_PRIORITY, to
allow configuring serial drivers separately from other devices. This is
similar to other driver classes like I2C and SPI.
The default is set to CONFIG_KERNEL_INIT_PRIORITY_DEVICE to preserve the
existing default initialization priority for most drivers. The one
exception is uart_lpc11u6x.c which previously used
CONFIG_KERNEL_INIT_PRIORITY_OBJECTS.
This change was motivated by an issue on the frdm_k64f board where the
serial driver was incorrectly initialized before the clock control
driver.
Signed-off-by: Maureen Helm <maureen.helm@intel.com>
Updated uart_rx_enable() and uart_tx() to use timeout given
in microseconds. Previously argument was given in milliseconds.
However, there are cases when milliseconds granularity is not
enough and can significantly reduce a throughput, e.g. 1ms is
100 bytes at 1Mb.
Updated 4 drivers which implement asynchronous API. Updated
places where API was used.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
async_rx_disable does not re-anable RXNE interrupt, it was disabled
during async_rx_enable
Signed-off-by: Manojkumar Subramaniam <manoj@electrolance.com>
Now that we're clearer around pm constraints management in various
TX cases (poll streams, irq driven or async), make some code
simplifications to ease readability.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
Introduce new logic to set/release pm_constraint during serial TX
transactions.
First change is to introduce an internal flag and utility functions
to control the set/release constraint balancing per uart device.
This way, whatever the mix of transactions or API calls, we
ensure a single uart device can only do 1 or 0 to the PM state
constraint. Constraint can't then be set more than once, released w/o
having been set or released more than it was set.
The last part of the change reworks the triggers for constraints
set/release operations.
In order not to disturb driver operations, if irq driven mode or PM is
enabled, don't enable TC interrupt handling by default.
Instead, map the pm_constraint setting to the way TC flag is handled
in normal mode of operations (irq driven or async).
As a consequence, in irq driven mode, pm_constraint is set/released on
tx_enable/tx_disable api calls, which gives API user full control
on transaction protection vs low power operations.
Finally, we emulate the same behavior on TX poll transaction, by
enabling TC irq at the start of a stream and disabling TC irq once
stream is completed. This is controlled with a dedicated device flag.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
According to reference manual, use of TC is "to avoid corrupting the
last transmission when the USART is disabled or enters Halt mode.". It
is safe to remove it since it is not checked when CONFIG_PM=n.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
In the current implementation the STM32 UART driver required to enable
`CONFIG_PM_DEVICE` when `CONFIG_PM=y` to function properly. The main
reason is that in some situations, like in polling mode, transmissions
are not fully synchronous. That is, a byte is pushed to the _queue_ if
it is empty and then the function returns without waiting for it to be
transmitted to the wire. This makes sense to make things like per-byte
transmission efficient. However, this introduces a problem: the system
may reach idle state, and so enter low power modes before the UART has
actually finished the last data in the queue. If this happens,
communications can be interrupted or garbage data may be put into the
UART line.
The proposed solution in this patch uses PM constraints to solve this
problem. For the IRQ/DMA case it is easy since we can set the constraint
before transmission start, and when the completion (TC) interrupt is
received we can clear it. However, the polling mode did not have the
capability to signal the completion. For this case, a simpler IRQ
routine is provided to just release the constraint. As a result, the PM
hooks are not required and so system can operate with just `CONFIG_PM`.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The STM32_DMA_SLOT macro from include/drivers/dma/dma_stm32.h
must be used here, especially for dma of type v2bis.
In this case, the dma-cell is not defined and slot is null.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
Instead of passing target states, use actions for device PM control.
Actions represent better the meaning of the callback argument.
Furthermore, they are more future proof as they can be suitable for
other PM actions that have no direct mapping to a state. If we compare
with Linux, we could have a multi-stage suspend/resume. Such scenario
would not have a good mapping when using target states.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
- Return -ENOTSUP if the requested state is not supported
- Remove redundant "noop style" functions.
- Use switch everywhere to handle requested state (not necessary in all
drivers, but better take off with consistency in place after current
changes).
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The device PM control function will only be called if the requested
state is different from the current one. A significant amount of drivers
were checking for state changes, now unnecessary. This patch removes all
this redundant logic.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Since the state is no longer modified by the device PM callback, just
use the state value.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The callback is now invoked to set the device PM state in all cases, so
the usage of ctrl_command is redundant.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The device PM subsystem already holds the device state, so there is no
need to keep duplicates inside the device. The pm_device_state_get has
been refactored to just return the device state. Note that this is still
not safe, but the same applied to the previous implementation. This
problem will be addressed later.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
All the macro for dma-cells are now in the
include/drivers/dma/dma_stm32.h header file.
So the include/dt-bindings/dma/stm32_dma.h is no more
useful and removed from #include.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
The macro to set the element of the dma-cells for each peripheral
are defined in the dma_stm32 header file
and used in the periph driver (as dma client)
Signed-off-by: Francois Ramu <francois.ramu@st.com>
The callback is not used anymore, so just delete it from the pm_control
callback signature.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
the device PM callback is not used anymore by the device PM subsystem,
so remove it from all drivers/tests using it.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Move all PM_DEVICE_STATE_* definitions to an enum. The
PM_DEVICE_STATE_SET and PM_DEVICE_STATE_GET definitions have been kept
out of the enum since they do not represent any state. However, their
name has not been changed since they will be removed soon.
All drivers and tests have been adjusted accordingly.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Since we removed various series headers, move stm32 driver
under main driver/pinmux folder.
Take this change into account into various drivers.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
This kconfig option enables runtime configuration of UART
controllers. This allows application to call uart_configure()
to configure the UART controllers and calling uart_config_get()
to retrieve configuration. If this is disabled, UART controllers
rely on UART driver's initialization function to properly
configure the controller. The main use of this option is mainly
code size reduction.
Fixes#16231
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
When using the uart driver with interrupt and async api
at the same time (instance for interrupt and instance for async),
the transmission complete interrupt was handled in the async
handling section, even when interrupt driven api is used.
This caused transmission to not work properly in interrupt mode.
The fix is to move the interrupt mode handling to the begginning
of the isr. If async mode is used then interrupt mode code
will not be run.
Signed-off-by: Shlomi Vaknin <shlomi.39sd@gmail.com>
Resetting the dma_tx.buffer_length after the dma_tx.counter
calculation, instead of before.
We need to reset the buffer_length when the transmission is finished,
but in order to give the correct value to the uart_event_tx struct,
we use the dma_tx.buffer_length in the calculation of the
dma_tx.counter, used for the len of the event (number of bytes sent).
I found this problem, when I wanted to use the uart_event_tx.len for
freeing the used space inside a ring buffer (ring_buf_get_finish),
and it didn't work, I logged the values of the uart_event_tx struct,
and found out it always was 0, because the buffer_length was 0,
and the whole buffer was transmitted (stat.pending_length also 0).
Signed-off-by: Prema Jonathan van Win <jonathanvanwin@gmail.com>
The context parameter used across device power management is
actually the power state. Just use it and avoid a lot of
unnecessary casts.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
