Replace slave_register with target_register and slave_unregister
with target_unregister.
Signed-off-by: Rajavardhan Gundi <rajavardhan.gundi@intel.com>
Signed-off-by: Jose Alberto Meza <jose.a.meza.arellano@intel.com>
With the device_sync_sem semaphore, there is the possibility of
the code not returning to give it back
(mcux_flexcomm_master_transfer_callback is never called),
causing it to get stuck in k_sem_take(&data->device_sync_sem, K_FOREVER)
in subsequent calls.
The i2c driver recovers by other means
(enabling FSL_FEATURE_I2C_TIMEOUT_RECOVERY)
but the callback might not return.
Adding a timeout option allows for this occurrence to be avoided.
Signed-off-by: Guilherme Casa Nova <guilherme.casa_nova@dell.com>
This fixes following build error when rv32m1_vega_zero_riscy
board gets compiled:
implicit declaration of function 'INST_DT_CLOCK_IP_NAME'
Fixes#68012
Signed-off-by: Armando Visconti <armando.visconti@st.com>
When adjusting the prescale to increase the I2C SCL low period,
the high period must also be subtracted to maintain a consistent
frequency.
Signed-off-by: Tim Lin <tim2.lin@ite.corp-partner.google.com>
When the I2C is in the target mode and encounters the bus error, the
driver has to reset the bus to recover the I2C hardware. However, when
the hardware is disabled, the interrupt enable bits are also cleared
automatically by design. As a result, we need to enable the interrupts
after resetting the I2C hardware.
Signed-off-by: Jun Lin <CHLin56@nuvoton.com>
This change marks each instance of the 'i2c_driver_api' as 'static const'.
The rationale is that 'i2c_driver_api' is used for declaring internal
module interfaces and is not intended to be modified at runtime.
By using 'static const', we ensure immutability, leading to usage of only
.rodata and a reduction in the .data area.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
Adding a hook for tests to inject a mock transport and migrating the
accel test to test bmi160 specific things. The old version of the test
which checks for read values is now covered by the generic test in
the sensor build_all target.
Signed-off-by: Yuval Peress <peress@google.com>
Employ a code spell checking tool to scan and correct spelling errors
in all files within the drivers/i2c directory.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
With commit 1d7476af, it fixed a build error with config
structure no longer having a base address field but left
the local variable defined. This resulted in a build warning
of an unused variable 'config' in two places.
Signed-off-by: David Leach <david.leach@nxp.com>
With commit 734adf52c6, the MCUX LPI2C config structure no longer
contains a direct base address pointer. The base address must be
accessed via DEVICE_MMIO_NAMED_GET. Some declarations in the LPCI2C
target mode handler still used the old method of accessing the base
address, causing a build failure. Fix these accesses to use the local
declaration of the "base" variable.
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
The controller can implement a reception FIFO as deep as 256 bytes.
However, the computation made by the driver code to determine how many
bytes can be asked is stored in a signed 8-bit variable called rx_empty.
If the reception FIFO depth is greater or equal to 128 bytes and the FIFO
is currently empty, the rx_empty value will be 128 (or more), which
stands for a negative value as the variable is signed.
Thus, the later code checking if the FIFO is full will run while it should
not and exit from the i2c_dw_data_ask() function too early.
This hangs the controller in an infinite loop of interrupt storm because
the interrupt flags are never cleared.
Storing the rx_empty empty on a signed 32-bit variable instead of a 8-bit
one solves the issue and is compliant with the controller hardware
specifications of a maximum FIFO depth of 256 bytes.
It has been agreed with upstream maintainers to change the type of the
variables tx_empty, rx_empty, cnt, rx_buffer_depth and tx_buffer_depth to
plain int because it is most effectively handled by the CPUs. Using 8-bit
or 16-bit variables had no meaning here.
Signed-off-by: Adrien Ricciardi <aricciardi@baylibre.com>
Enable runtime mmio configuration in mcux_lpi2c driver. The fsl_lpi2c
driver relies on physical address to determine instance number.
So mmap flag 'K_MEM_DIRECT_MAP' is required.
Signed-off-by: Chekhov Ma <chekhov.ma@nxp.com>
- Allow non RTIO i2c drivers to be intermixed with RTIO drivers
- Remove reference to rtio_spsc_drop_all()
- Fix impllicit cast which throws an error with more restrictive
compile flags
Signed-off-by: Yuval Peress <peress@google.com>
During init i2c_ambiq device, the bitrate calculation is not correct,
results in incorrect device speed, or failed to configure i2c device
if clock-frequency is set to higher than I2C_BITRATE_STANDARD
Signed-off-by: Bryan Zhu <bzhu@ambiq.com>
Fixes the routing of the events associated with a secondary
i2c target address being routed to the primary config. The
i2c_target_config/slave_cfg was being selected from the
driver address match but then over written by the primary.
This change fully implements the if/else of 10bit addressing
and includes a assert if the slave_cfg is NULL, and explains
why dual 10bit addresses on STM32 won't work.
Signed-off-by: Tim Woolliscroft <tim@opteran.com>
Do not enable subsystem/driver shell modules by default and stop abusing
CONFIG_SHELL_MINIMAL, which is internal to the shell subsystem, to decide
when to enable a driver shell.
The list of shell modules has grown considerably through the
years. Enabling CONFIG_SHELL for doing e.g. an interactive debug session
leads to a large number of shell modules also being enabled unless
explicitly disabled, which again leads to non-negligible increases in
RAM/ROM usage.
This commit attempts to establish a policy of subsystem/driver shell
modules being disabled by default, requiring the user/application to
explicitly enable only those needed.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Use GPIO_OUTPUT_INACTIVE to initialize the pin so that the ACTIVE_LOW DT
flag is honored and use the gpio_pin_set_dt functions to set the
(logical) value of the pin instead of gpio_pin_configure_dt, that
tries to reconfigure the pin each time.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
We need to do a build assert for the fifo enable status of 'I2C2'.
There is a problem with using instance to obtain property when
any one I2C port is not enabled.
Signed-off-by: Tim Lin <tim2.lin@ite.corp-partner.google.com>
Add drivers for gpio_i2c_switch which is present in beagleconnect freedom
Signed-off-by: Vaishnav Achath <vaishnav@beagleboard.org>
Signed-off-by: Ayush Singh <ayushdevel1325@gmail.com>
Add a property for I2C channel switch selection. This property will
write to the SMBxxCHS register according to the I2C node you selected,
which can make channel swapping.
Signed-off-by: Tim Lin <tim2.lin@ite.corp-partner.google.com>
The ISR for CQ mode was unconditionally writing into the I2C message
buffer for all transfer types. The correct data was transferred on the
I2C interface, but the clobber caused CONFIG_I2C_LOG_LEVEL_DBG to
display incorrect data for I2C writes.
This change will also help performance of large I2C write transactions.
Signed-off-by: Keith Short <keithshort@google.com>
Move the syscall_handler.h header, used internally only to a dedicated
internal folder that should not be used outside of Zephyr.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Add the get_config API to the stm32 I2C driver.
It will return the value of the Clock Control register for i2C V1 bus
or the TIMING register for the I2C V2 bus.
This is hold by a i2c_config_timing structure of the device data
Add a bool to check if the I2C is configured or not.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
Let the caller decide what should be done in case of the transfer failure.
It will reduce the number of error log prints when the i2c scan is
called for the bus where nothing is connected.
Signed-off-by: Bartosz Bilas <b.bilas@grinn-global.com>
In driver runtime configuration function, calling
calling pm_device_runtime_get/put() will have the effect of performing
pinctrl change from sleep to default and back from default to sleep
which is useless and in turn enables and disables GPIO clocks two times.
Stop this crazyness and purely enable/disable clock, which might be
superfluous in some cases but which remains much more reasonable than
than the previous implementation.
Signed-off-by: Erwan Gouriou <erwan.gouriou@st.com>
pm_device_runtime_enable() will call i2c_stm32_suspend() if
pm_device_init_suspended() isn't called.
Since the aim is to perform suspension, just need to call
pm_device_runtime_enable().
Signed-off-by: Erwan Gouriou <erwan.gouriou@st.com>
During i2c_configure API execution, there is no way core
can go in low power mode. Hence, call to pm_device_busy_set/get API
is useless.
Signed-off-by: Erwan Gouriou <erwan.gouriou@st.com>
i2c_rcar.c includes soc.h header which doesn't need for this source
and exists not for all boards. soc.h header doesn't exist for rcar-gen3
soc based on arm64 core.
soc.h consists soc-depended defenitions and need to be included
by soc-depended sources.
Signed-off-by: Dmytro Semenets <dmytro_semenets@epam.com>
Fixes an issue that reload mode is not disabled in case of an error.
From this case the driver could not recover because in msg_init()
no new transfer could be initialized.
Signed-off-by: Mario Jaun <mario.jaun@gmail.com>
Different nRF52 devices have different maximum TWI DMA transfer size,
and it's easy to hit the limit with i2c displays on nrf52832 (8 bit) and
nrf52810 (10 bit). Currently neither the driver or the hal validate the
limit, leading to random NACK errors when trying to transfer more data.
Add a check on the driver to fail gracefully when going over the limit.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
I2C scan might fail as peripheral is still busy completing last
operation. This makes sure transfer call waits for free line.
Signed-off-by: Sylvio Alves <sylvio.alves@espressif.com>
