Only initialize the FlexCAN IP core once since initialzing it has the
side effect of resetting the IP core and thus clearing previous settings
such as RX filters.
Fixes: #44680
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Move the code for socket instanciation from each driver
to a generic driver, that makes an instance of a socketCAN
net device for the chosen node.
Signed-off-by: Alexander Wachter <alexander@wachter.cloud>
Include a pointer to the CAN controller device for the CAN
transmit, receive, and state change callback functions.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
The CAN_BUS_UNKNOWN CAN controller state is only used to indicate that
the current CAN controller state could not be read.
Remove it and change the signature of the can_get_state() API function
to return an integer indicating success or failure.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Rework the transmit error handling in the NXP MCUX FlexCAN driver:
- Frame transmission must be automatically retried in case of lost
arbitration or missing acknowledge.
- Abort any pending TX frames when bus-off state is entered.
- Fail early in can_send() if in bus-off state.
Fixes: #19502
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Add optional user data argument to can_set_state_change_callback() to
comply with Zephyr API design guidelines.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Rename a few CAN API functions for clarity and consistency with other
Zephyr RTOS APIs.
CAN_DEFINE_MSGQ() becomes CAN_MSGQ_DEFINE() to match K_MSGQ_DEFINE().
can_attach_isr() becomes can_add_rx_filter() since a filter callback
function is not an interrupt service routine (although it is called in
isr context). The word "attach" is replaced with "add" since filters are
added, not attached. This matches the terminology used is other Zephyr
APIs better.
can_detach() becomes can_remove_rx_filter() to pair with
can_add_rx_filter().
can_attach_msgq() becomes can_add_rx_filter_msgq() and documentation is
updated to mention its relationship with can_add_rx_filter().
can_register_state_change_isr() becomes can_set_state_change_callback()
since a state change callback function is not an interrupt service
routine (although it is called in isr context). The word "register" is
replaced with "set" since only one state change callback can be in
place.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Deprecate the use of CAN-specific error return values and replace them
with standard errno values.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Catch up on the CAN driver API argument naming changes:
- Unify naming of callback function pointers as "callback".
- Unify naming of user-specified callback function arguments as
"user_data".
- Instances and pointers to struct zcan_frame are named "frame",
not "msg", to avoid confusion with the CAN message queue support.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Trivial fix of compilation error "invalid operands to binary "
when CONFIG_CAN_AUTO_BUS_OFF_RECOVERY = n
Fixes#40290
Signed-off-by: Lucas Dietrich <ld.adecy@gmail.com>
The maximum number of concurrent filters depends on the hardware and
the driver implementation. This API allows the application to obtain
the maximum number of available filters.
Signed-off-by: Martin Jäger <martin@libre.solar>
Refactors all of the CAN drivers to use a shared driver class
initialization priority configuration, CONFIG_CAN_INIT_PRIORITY, to
allow configuring CAN 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 exception
is the mcp2515 driver which has a dependency on a SPI driver and must
therefore initialize later than the default device priority.
Signed-off-by: Maureen Helm <maureen.helm@intel.com>
Fix the limits for the timing parameter calculations.
The lower limit for the phase_seg2 value is wrongly specified as 1 to 7,
but 1U is substracted before writing it to the CTRL1:PSEG2 register
field. This results in register field values between 0 and 6, but 0 is
an invalid value for the PSEG2 register field.
The upper limits for several of the timing parameters are wrong as well,
but this does not result in invalid register field values being
calculated. It can, however, result in not being able to meet CAN timing
requirements.
The confusion in specifying the limits likely stems from the timing
calculations and timing limits using the "physical" values, whereas the
registers fields all use the "physical" value minus 1. When the
datasheet says "The valid programmable values are 1-7", the
corresponding limits should be set to 2 to 8 to take the "minus 1" into
account.
Fixes: #39541
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
IRQ_CONNECT macro last argument is architecture-specific flag,
but now there is instance id at this place.
This PR set architecture-specific flag to 0.
Signed-off-by: Grixa Yrev <GrixaYrev@yandex.ru>
The RT1170 platform adjusted the result parameter in the callback
function to be a uint64_t. The adjustment is to conditionally change
the callback definition and then cast the result value to a size that
fits the reported values across the various SOCs.
With error results, cast result to uint64_t and modified the called
function to print a uint64_t value.
With the RX/TX status, result is a MB value that doesn't exceed 64
so the result value is cast down to a uint32_t.
Fixes#37691
Signed-off-by: David Leach <david.leach@nxp.com>
If the supplied sjw in the timing parameters is zero,
the sjw parameter should not be changed.
This fixes the uninitialized swj in can_set_bitrate.
Signed-off-by: Alexander Wachter <alexander@wachter.cloud>
socket_can_generic.h makes some assumptions that are not always valid
for various drivers with regards to numbering. To clean this up we
add defines for SOCKET_CAN_NAME_0 and SOCKET_CAN_NAME_2 in addition to
SOCKET_CAN_NAME_1.
We also move struct socket_can_context into the drivers themselves
since they know best how many CAN interfaces are getting supported and
what naming/number convention they'd have.
Additionally, this also exposes a few other build issues with the
can_mcux_flexcan driver that get fixed.
Finally, we remove the platform_allow from samples/net/sockets/can
since it is no longer needed.
Fixes#31733
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Check if the timing is not set from device tree and exclude
sample-point algorithm if it is not used from device-tree.
Signed-off-by: Alexander Wachter <alexander@wachter.cloud>
Reordering of the struct elements to match the Linux format.
The __packed() is not necessary anymore.
std_id and ext_id is merged to id in the frame and filter.
Additionally, the frames are ready for CAN-FD.
Signed-off-by: Alexander Wachter <alexander@wachter.cloud>
The previous API can't change the sampling-point and only allowed
bitrates that fit the time segments.
The new API allows for shifting the sampling-point and adjusts the
number of time quantum in a bit to all more possible bitrates.
The functions to calculate the timings are moved to the can_common file.
They can be used for all drivers.
Signed-off-by: Alexander Wachter <alexander@wachter.cloud>
Adds imx rt support.
Allows n-number of can interfaces based on device-tree.
Adds a "common" irq name.
Added CAN bus pins and dts for 1060 and 1064 EVK.
Signed-off-by: Rick Talbott <rtalbott@fastmail.com>
It is necessary to wrap the device pointer into data, and pass the data
to the HAL callback function.
Fixes#27399
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Now that device_api attribute is unmodified at runtime, as well as all
the other attributes, it is possible to switch all device driver
instance to be constant.
A coccinelle rule is used for this:
@r_const_dev_1
disable optional_qualifier
@
@@
-struct device *
+const struct device *
@r_const_dev_2
disable optional_qualifier
@
@@
-struct device * const
+const struct device *
Fixes#27399
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
-Wimplicit-fallthrough=2 requires a fallthrough comment or a compiler
to tells gcc that this happens intentionally.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Several reviewers agreed that DT_HAS_NODE_STATUS_OKAY(...) was an
undesirable API for the following reasons:
- it's inconsistent with the rest of the DT_NODE_HAS_FOO names
- DT_NODE_HAS_FOO_BAR_BAZ(node) was agreed upon as a shorthand
for macros which are equivalent to
DT_NODE_HAS_FOO(node) && DT_NODE_HAS_BAR(node) &&
- DT_NODE_HAS_BAZ(node), and DT_HAS_NODE_STATUS_OKAY is an odd duck
- DT_NODE_HAS_STATUS(..., okay) was viewed as more readable anyway
- it is seen as a somewhat aesthetically challenged name
Replace all users with DT_NODE_HAS_STATUS(..., okay), which is
semantically equivalent.
This is mostly done with sed, but a few remaining cases were done by
hand, along with whitespace, docs, and comment changes. These special
cases include the Nordic SOC static assert files.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Fix use of DT_HAS_DRV_INST which does not exist.
Use DT_HAS_NODE_STATUS_OKAY(DT_DRV_INST(n)) instead.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
As we phase out per instance Kconfig symbols convert to utilize
DT_DRV_INST to initialize CAN instances.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Convert all canbus related API/samples/tests/subsys
to the new timeout API with k_timeout_t.
Signed-off-by: Alexander Wachter <alexander@wachter.cloud>
By changing the various *NET_DEVICE* macros. It is up to the device
drivers to either set a proper PM function or, if not supported or PM
disabled, to use device_pm_control_nop relevantly.
All existing macro calls are updated. Since no PM support was added so
far, device_pm_control_nop is used as the default everywhere.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
