This only fix several test cases failed while running code coverage
report inmps2_an385 platform. Enlarge the stack size for which failed
due to MPU fault of stack overflow.
Signed-off-by: Enjia Mai <enjiax.mai@intel.com>
If Zephyr is running on a coprocessor we might lack I/O
such as uart or usb to output tracing datas but we might
have gigabytes of RAM available.
This patch allows to output trace datas to a ram buffer, which then
may be retrieved using gdb.
e.g:
(gdb) dump binary memory channel0_0 <ram_tracing_start> \
<ram_tracing_end>
Signed-off-by: Julien Massot <julien.massot@iot.bzh>
In C, `void foo(void);` and `void foo();` mean subtly different things.
The former means "foo takes zero arguments"; the latter means
"foo takes an unspecified number of arguments". This can result in
calling convention mismatches in exceptional cases.
Change to emitting `(void)` instead of `()` for a syscall with
zero arguments.
Signed-off-by: James Harris <james.harris@intel.com>
There was an error in the ordering of the parameters in the
DEVICE_DT_DEFINE for the native POSIX counter. This made a project
using a counter built for native posix not being able to compile.
This commit switches places for ctr_init and device_pm_control_nop.
Signed-off-by: Tofik Sonono <tofik@sonono.me>
add CAN_1 and CAN_2 hw modules support
Signed-off-by: Alexander Kozhinov <AlexanderKozhinov@yandex.com>
Co-authored-by: Erwan Gouriou <erwan.gouriou@linaro.org>
Currently there is no way to distinguish between a caller
explicitly asking for a semaphore with a limit that
happens to be `UINT_MAX` and a semaphore that just
has a limit "as large as possible".
Add `K_SEM_MAX_LIMIT`, currently defined to `UINT_MAX`, and akin
to `K_FOREVER` versus just passing some very large wait time.
In addition, the `k_sem_*` APIs were type-confused, where
the internal data structure was `uint32_t`, but the APIs took
and returned `unsigned int`. This changes the underlying data
structure to also use `unsigned int`, as changing the APIs
would be a (potentially) breaking change.
These changes are backwards-compatible, but it is strongly suggested
to take a quick scan for `k_sem_init` and `K_SEM_DEFINE` calls with
`UINT_MAX` (or `UINT32_MAX`) and replace them with `K_SEM_MAX_LIMIT`
where appropriate.
Signed-off-by: James Harris <james.harris@intel.com>
Remove stale references to v2m_musica in the ipc samples as the board
support for v2m_musca has been removed.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Maximum GIC SPI is 228, so maximum INTID is 228 + 32 = 260.
So, _sw_isr_table needs to support 260 entries
Signed-off-by: Raveendra Padasalagi <raveendra.padasalagi@broadcom.com>
Signed-off-by: Abhishek Shah <abhishek.shah@broadcom.com>
Update register addresses and NVIC/GIC interrupt numbers
according to the latest viper RTL version.
Signed-off-by: Raveendra Padasalagi <raveendra.padasalagi@broadcom.com>
Signed-off-by: Abhishek Shah <abhishek.shah@broadcom.com>
If a shell command is compiled out using SHELL_COND_CMD(),
a line for this command will still be printed but will
be blank. Change it so compiled out commands are not
listed as blank lines.
Signed-off-by: Pete Skeggs <peter.skeggs@nordicsemi.no>
Added CLOCK_CONTROL_NRF_FORCE_ALT dependency to some options which
are not valid when clock is controlled by out-of-tree driver.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Modified CLOCK_CONTROL_NRF_ACCURACY to represent integer value of
LF clock accuracy.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
The test case suggests that "For most arch which support
userspace, derefencing NULL pointer will be caught by
exception.". This is certainly not true for Cortex-M, where
read access to address 0x0 is generally allowed. The reason
the test had been passing was either 1) because in many
Cortex-M platforms, including QEMU, address 0x0 is unmapped,
or 2) GCC is generating an undefining instruction. However,
now that we have activated the null pointer derefrencing
detection, we may end up with two exceptions and the test
would fail.
Change illegal access to something outside the mapped memory
area, e.g. 0xFFFFFFFF.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
In the arch/arm test suite there are tests running with
CONFIG_NO_OPTIMIZATIONS enabled, and may require a
larger idle thread stack size.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Temporarily save and restore all callee-saved registers
in the sub-routine, not only the frame pointer, to avoid
errors if compiler chooses to use them in the alternative
thread function. The callee-saved regs are restored after
the alternative thread switches back in.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
The shim layer could in some circumstances not be properly
configured which would result in an unbound radio interrupt
handler.
Signed-off-by: Rafał Kuźnia <rafal.kuznia@nordicsemi.no>
This commit moves all hal_nordic radio driver code that is
strictly dependent on Zephyr into the Zephyr repository.
Signed-off-by: Rafał Kuźnia <rafal.kuznia@nordicsemi.no>
Its been 2 releases since we deprecated the old zephyr integer types.
We can now remove support for the Kconfig option, code, and, test to
allow the old int types to be supported.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Add NET_HEXDUMP_DBG/ERR/WARN/INFO macros, then use them for new
MQTT_HEXDUMP_TRC/ERR/WARN/INFO macros.
Log struct mqtt_utf8 using MQTT_HEXDUMP_TRC. One cannot safely log
mqtt_utf8 strings due to no guarantee of a NULL terminator being
present. Also, logging without log_strdup() as if it were a NULL
terminated string asserts when CONFIG_LOG_IMMEDIATE=n. This solves
both issues.
Signed-off-by: Pete Skeggs <peter.skeggs@nordicsemi.no>
Tests to exercise the new `cbprintf_package()`, `cbvprintf_package()`
and `cbpprintf()`.
[ Heavily based on a prior proposal from Peter Bigot. ]
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
In applications like logging the call site where arguments to
formatting are available may not be suitable for performing the
formatting, e.g. when the output operation can sleep. Add API that
supports capturing data that may be transient into a buffer that can
be saved, and API that then produces the output later using the
packaged arguments.
[ Documentation and commit log from Peter Bigot. ]
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Due to the recent changes to scheduler z_find_first_thread_to_unpend
& z_remove_thread_from_ready_q are not used anymore. So removing the
dead code.
fixes: #32691
Signed-off-by: Spoorthy Priya Yerabolu <spoorthy.priya.yerabolu@intel.com>
Any value passed to a function in <ctype.h> shall be
representable as an unsigned char or be the value EOF.
So changed type of variable to unsigned char.
Signed-off-by: Spoorthy Priya Yerabolu <spoorthy.priya.yerabolu@intel.com>
Uses of k_work_pending are to be replaced by k_work_is_pending which
conforms to current proposed naming guidelines.
Switch to the new function.
Also initialize the private work structure at build time, rather than
on each iteration (it is not permitted to invoke work API on an
uninitialized work item).
The implementation here is racy: that a work item is pending does not
mean changes since it was first submitted are guaranteed to be seen
when the work item begins (began) executing.
A better solution would be to have transmit_message be able to
determine whether there is unprocessed work. Then the work item can
be submitted unconditionally.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Uses of k_work_pending are to be replaced by k_work_is_pending which
conforms to current proposed naming guidelines.
Both uses in this file are fragile: that a work item is pending does
not mean changes since it was first submitted are guaranteed to be
seen when the work item begins (began) executing.
As long as this module is expected to be replaced by tcp2 it doesn't
seem worth trying to fix the logic, so just switch to the new function
name.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Checking whether a work item is pending before submitting it is racy:
the item may be finishing up in its handler, and essentially
completed, in which case chosing not to resubmit would leave work
unhandled.
In this case it appears very wrong, since the sole call site in
net_if.c has just initialized the work item, which is not permitted if
the work item is pending.
Remove the check.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Add Thalley (emil.gydesen@nordicsemi.no) and asbjornsabo
(asbjorn.sabo@nordicsemi.no) as additional maintainers/codeowners for
bluetooth audio.
(Also adds empty audio.h file, that will later be expanded, so that
the include/bluetooth/audio directory now listed in CODEOWNERS
actually exists.)
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
Improve dynamic interrupt test cases of interrupt for platform such as
x86, x86_64, native_posix, this improve code coverage of it.
Signed-off-by: Enjia Mai <enjiax.mai@intel.com>
Enable the CONFIG_TEST switch in the test suite.
This allows certain required Kconfig settings for
ARM Cortex-M (which depend on TEST) to be enabled
when executing the test suite.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
The compiler generates a warning regarding a variable being used
w/o being initialized in certian configuration. According to the logic
that's not the case, so just add some initial value to the variable to
silence the compiler.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
As network interface array size might be larger than the actual
network interface count, check this condition and ignore those
interfaces that are not in use.
We cannot know for certain how many network interfaces there
are at built time, as the total count is only available at runtime.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The network interface index is very useful info to see in
debug prints so add those to debug output.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
While I'm in the idle code, let's clean this loop up. It was a really
bad #ifdef hell:
* Remove the CONFIG_TICKLESS_IDLE_THRESH logic (and the kconfig),
which never did anything but needlessly increase latency.
* Move the needed timeout logic from the main loop into
pm_save_idle(), which eliminates the special case for
!SYS_CLOCK_EXISTS.
Behavior (modulo that one kconfig) should be completely unchanged, and
now the inner part of the idle loop looks like:
while (true) {
(void) arch_irq_lock();
if (IS_ENABLED(CONFIG_PM)) {
pm_save_idle();
} else {
k_cpu_idle();
}
}
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
This kconfig isn't actually exercised in this test, it's just being
used to compute some sleep durations. Also I want it gone.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The removal of the abort handling also absconded with an IRQ lock that
is required for reliable operation in the idle loop. Put it back.
Once the idle loop has made a decision to enter idle, any interrupt
that arrives needs to be masked and delivered AFTER the system enters
idle. Otherwise we run the risk of races where the system accepts and
processes an interrupt that should have prevented idle, but then goes
to sleep anyway having already made the decision.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
CPython is sometimes described as "single threaded" due to the GIL,
but the interpreter will still "preemptively" switch between threads
(the details seem poorly documented).
So the time between checking whether acquire is 1 and decrementing the
count could result in more than one thread seeing an "available"
device, and more than one test being run (simultaneously, on the same
physical device!). We have a big herd of threads all polling for
this, so in a large test run this would happen maybe one time out of
20-30 tries.
Use a lock. Also remove the very similar looking
DUT.get_available_device() method, which had the same bug but appears
to be dead code.
Fixes#32679
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
This patch adds weak sys_arch_reboot() function to avoid build error
with CONFIG_REBOOT=y. Some SoC has already had own reboot function
but others (Ex. qemu boards) faced buld error.
- openisa_rv32m1: Not change
- riscv-ite: Do nothing, remove and use arch/riscv function
Signed-off-by: Katsuhiro Suzuki <katsuhiro@katsuster.net>
Instead of hardcoding the network interface value, use
net_if_ipv6_select_src_iface() so that proper network interface
gets used.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
If we are sending a network packet and if the remote address
is not set in the context (which means that connect() has not
been called), then we must set the target network interface
to a proper value.
This is done so that when we select the local source address,
we might select the wrong interface if we have multiple network
interfaces in the system. In this case the packet would be always
assigned to first network interface regardless of the destination
address.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Remove unnecessary main stack configuration in bluetooth samples.
The HCI power control sample does not need to set the main stack this
low, and setting it causes restriction on what the application can
change in menuconfig. A change in the entropy driver causes stack
overflow.
The HCI uart sample sets the main stack to the default size.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
If files are changed for architecture code, then only run CI on those
architectures.
Combine all scripts into one and make it simpler.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
