Previously it was not possible to n-select e.g.
CONFIG_PTHREAD_BARRIER because the Kconfig template for it
lacked a prompt. This made it impossible to disable some
unused POSIX features and unnecessarily increased code size
if unused code sections were not discarded by the linker.
Add a prompt so that each pooled IPC type is
user-configurable and can be disabled if unneeded.
Further, ensure that only the selected sources from lib/posix
are included in the cmake build.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Ensure that stdin, stdout, and stderr are initialized statically.
Previously, the mutex and condition variable were uninitialized.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Leave the malloc partition so that it only contains the heap itself; this
lets the initialization code adjust the address range when configuring the
arena at startup.
Signed-off-by: Keith Packard <keithp@keithp.com>
Instead of adding every possible subsystem which places variables in the C
library memory partition in libc-hooks.h, place those conditions in the
related Kconfig files and simplify the libc-hooks.h to just looking at
CONFIG_NEED_LIBC_MEM_PARTITION.
Signed-off-by: Keith Packard <keithp@keithp.com>
rand32.h does not make much sense, since the random subsystem
provides more APIs than just getting a random 32 bits value.
Rename it to random.h and get consistently with other
subsystems.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Picolibc dependencies limit ability to use third party minimal
implementations of CPP when enablng PICOLIBC_USE_MODULE.
Signed-off-by: Al Semjonovs <asemjonovs@google.com>
Fixes#62677 where printf defaults to using armclang's semihosting backed
implementation of printf.
Signed-off-by: Wilfried Chauveau <wilfried.chauveau@arm.com>
to_posix_mutex allocates a mutex which will race to change the value of
the lock, thus to_posix_mutex should be performed under the spinlock to
prevent the racy issue.
Signed-off-by: Jaxson Han <jaxson.han@arm.com>
Zephyr's code base uses MP_MAX_NUM_CPUS to
know how many cores exists in the target. It is
also expected that both symbols MP_MAX_NUM_CPUS
and MP_NUM_CPUS have the same value, so lets
just use MP_MAX_NUM_CPUS and simplify it.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Integrates object core statistics framework into the following
kernel objects:
sys_mem_blocks, k_mem_slab
threads, _cpu, z_kernel
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
Rearranges the sys_mem_blocks fields so that information that describes
how much of the memory block is used is co-located. This will allow
easier of its statistics into the object core statistics reporting
framework.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
Adds the linkable loadable extensions (llext) subsystem which provides
functionality for reading, parsing, and linking ELF encoded executable
code into a managed extension to the running elf base image.
A loader interface, and default buffer loader implementation,
make available to the llext subsystem the elf data. A simple management
API provide the ability to load and unload extensions as needed. A shell
interface for extension loading and unloading makes it easy to try.
Adds initial support for armv7 thumb built elfs with very specific
compiler flags.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Co-authored-by: Chen Peng1 <peng1.chen@intel.com>
Co-authored-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
The original idea of z_current_get() was to be the counterpart
of k_current_get() when thread local variable for current has
not been initialized if TLS is enabled, otherwise they are
the same function. Now since z_current_get() is being used
outside of core kernel, rename it under kernel namespace so
other subsystem can conceptually use them too.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The intent of this patch is to leave all of the semantics of the macros
unchanged, only replacing the easy-to-read static inline conversion
function with a pile of ?: operators.
Ick. This is not a cleanup. However, what it does enable is using constant
results while initializing global variables, which cannot be done with
either static inline functions or even statement expressions, even when
those generate constant results.
Signed-off-by: Keith Packard <keithp@keithp.com>
Add two new functions: crc4 for generic calculations
of CRC4, and crc4_ti which use look-up table for
faster calculations of CRC4 algortihms that base
on 0x03 polynomial.
Signed-off-by: Michal Morsisko <morsisko@gmail.com>
When multithreading is disabled, the whole mutex infrastructure isn't
available. The common malloc code wasn't checking for this case which
caused build failures.
Signed-off-by: Keith Packard <keithp@keithp.com>
Tested-by: Detlev Zundel dzu@member.fsf.org
Log modules should be registered with LOG_MODULE_REGISTER
rather than LOG_MODULE_DECLARE. It seems the latter works
on most platforms (at least with in minimal mode as configured
with ZTest).
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Support logging for all POSIX pooled resource types such as
pthread_t, pthread_mutex_t, and pthread_cond_t.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
If `pthread_exit()` is called from a `k_thread`, then we would
previously trigger an assertion. The problem with that, is that
is POSIX is acting as a compatibility layer.
Given that it is a reasonable expectation to have the calling
thread exit or abort when calling `pthread_exit()`, lets do just
that.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
...when toolchain does not have native picolibc support.
Without native support, picolibc needs to be built from
the module. Disabling the prompt means this kconfig
takes on whatever default value specified, which is
to build source from module when there is no native
support.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The picolibc kconfigs were tied to ZEPHYR_TOOLCHAIN_VARIANT
and it only supported the Zephyr SDK, which is the only
toolchain having picolibc built-in at the moment. This
commit generalizes that to use TOOLCHAIN_HAS_PICOLIBC
cmake variable the same way as newlib: TOOLCHAIN_HAS_NEWLIB.
This provides the ability for other toolchains to declare
their support for picolibc.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This header does not expose any public APIs, so move it under
kernel/include and change files including it.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
add support for retrieve MAD and DMAR table information. Provided
two new interface namely acpi_dmar_entry_get() and acpi_drhd_get()
for retrieve DMA Remapping Reporting and DMA-remapping hardware
unit definition (DRDH).
Signed-off-by: Najumon B.A <najumon.ba@intel.com>
Modify the signature of the k_mem_slab_free() function with a new one,
replacing the old void **mem with void *mem as a parameter.
The following function:
void k_mem_slab_free(struct k_mem_slab *slab, void **mem);
has the wrong signature. mem is only used as a regular pointer, so there
is no need to use a double-pointer. The correct signature should be:
void k_mem_slab_free(struct k_mem_slab *slab, void *mem);
The issue with the current signature, although functional, is that it is
extremely confusing. I myself, a veteran Zephyr developer, was confused
by this parameter when looking at it recently.
All in-tree uses of the function have been adapted.
Fixes#61888.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Implements the posix clock_nanosleep function, where both relative
and absolute sleeps are made as absolute sleeps.
The nanosleep() function is a special case of clock_nanosleep(),
and so has been refactored to simply call it.
Signed-off-by: Tom Finet <tom.codeninja@gmail.com>
add explicit boolean type to 'if' statement controlling expression, thus
improving code readability and maintainability, complying with required
[misra-c2012-14.4] rule which states; The controlling expression of an
if statement and the controlling expression of an iteration-statement
shall have essentially boolean type.
Found as a coding guideline violation (Rule 14.4) by static code
scanning tool.
Note: Tested on STM32L5 Nucleo-144 board (stm32l552xx).
Signed-off-by: ferar alashkar <ferar.alashkar@gmail.com>
* `struct sigevent` is not type-defined
* `union sigval` is not type-defined
* `struct sigevent` must include `sigev_notify_attributes`
For more information, see https://bit.ly/3YfnELI
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
This enables -Wshadow to warn about shadow variables on
in tree code under arch/, boards/, drivers/, kernel/,
lib/, soc/, and subsys/.
Note that this does not enable it globally because
out-of-tree modules will probably take some time to fix
(or not at all depending on the project), and it would be
great to avoid introduction of any new shadow variables
in the meantime.
Also note that this tries to be done in a minimally
invasive way so it is easy to revert when we enable
-Wshadow globally. Source files under modules/, samples/
and tests/ are currently excluded because there does not
seem to be a trivial way to add -Wshadow there without
going through all CMakeLists.txt to add the option
(as there are 1000+ files to change).
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Memalign is another name for the posix aligned_alloc function, although it
has weaker restrictions on the relationship between the alignment and size.
memalign() is used internally by the libstdc++ when built for 'newlib'
targets (which includes picolibc) instead of aligned_alloc() due to a bug
in gcc, so we need to provide an implementation of this when using that
library, even though it's not part of the Zephyr C library API.
When a fix for the libstdc++ is merged upstream and can be consider a
reasonable dependency for Zephyr, this work-around can be removed.
Closes: #57899
Signed-off-by: Keith Packard <keithp@keithp.com>
Add new option to use thread local storage for stack
canaries. This makes harder to find the canaries location
and value. This is made optional because there is
a performance and size penalty when using it.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
When aiming for the smallest build result, it is desirable to
use CONFIG_CBPRINTF_NANO together with CONFIG_MINIMAL_LIBC. However
that doesn't mean we need all the functions enabled by having
CONFIG_CBPRINTF_LIBC_SUBSTS=y which increases the binary size even if
those functions are not used and not linked in.
When CONFIG_CBPRINTF_LIBC_SUBSTS=n such functions are not defined
anyway so z_cbvprintf_impl() won't be invoked with missing
functionalities.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
extends the char buffer in the strsignal function to cover the
entire range of `int`
Had to use `-INT_MAX` as the compiler resolves
```
STRINGIFY(INT_MIN)
```
to:
```
(-2147483647 - 1)
```
instead of:
```
-2147483648
```
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Add a new API to perform an immediate system power off:
`sys_poweroff()`.
Until now, this functionality has been implemented via the system power
management module, but in a clunky fashion. The way system PM works is
by defining some idle states in devicetree, that, given some properties
(e.g. minimal residency, exit latency, etc.) are automatically selected
when system goes to idle based on the expected next wake-up. However,
system off is a power state that one typically wants to control manually
from the application because it implies state loss, and in most cases,
configuring some sort of wake-up source. So in general, it is not
desired to let the system enter this state automatically. This led to
the following stuff in-tree:
from `boards/arm/mimxrt595_evk/mimxrt595_evk_cm33.dts`:
```c
/*
* Deep power-down mode is supported in this SoC through
* 'PM_STATE_SOFT_OFF' state. There is no entry for this in device tree,
* user can call pm_state_force to enter this state.
*/
```
That is, state not being defined in devicetree so that PM subsystem
doesn't pick it automatically, but still implemented in in the PM hooks:
from `soc/arm/nxp_imx/rt5xx/power.c`, `pm_state_set()`:
```c
case PM_STATE_SOFT_OFF:
set_deepsleep_pin_config();
POWER_EnterDeepPowerDown(EXCLUDE_FROM_DEEP_POWERDOWN);
break;
```
And to actually make use of this state, users had to do this kind of
abominations:
```c
pm_state_force(0u, &(struct pm_state_info){ PM_STATE_SOFT_OFF, 0, 0 });
/* Now we need to go sleep. This will let the idle thread runs and
* the pm subsystem will use the forced state. To confirm that the
* forced state is used, lets set the same timeout used previously.
*/
k_sleep(K_SECONDS(SLEEP_S));
printk("ERROR: System off failed\n");
while (true) {
/* spin to avoid fall-off behavior */
}
```
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
This patch adds Kconfig options to select either GNU libgcc or LLVM
compiler-rt. The 'rtlib' flag is provided in a config file, so this
patch introduces 'clang_libgcc.cfg' and 'clang_compiler_rt.cfg' which
enable appropriate library. The file is selected by concatenating
the 'clang_' prefix with library name.
Signed-off-by: Patryk Duda <pdk@semihalf.com>
The normative spec for `pthread_mutex_timedlock()` says that
it should return `ETIMEDOUT` when a timeout occurs. However,
currently it returns `EAGAIN`, which reflects what is returned
by `k_mutex_lock()`.
Inspect and update the return value to account for this slight
difference.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Picolibc doesn't need these two syscall implementations as it uses
zephyr_fputc instead. Make sure that zephyr_putc is declared correctly.
Signed-off-by: Keith Packard <keithp@keithp.com>
Ensure that the thread return value is set by `pthread_join()`
when `status` is non-NULL.
Additionally, we have an opportunity to synchronously clean
up thread stacks in `pthread_join()`, which is preferable.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Discovered this while implementing c11 threads, but there
was a regression recently that made it so that `pthread_join()`
would report success when attempting to join a thread that had
been detached with `pthread_detach()`.
Technically now that is undefined behaviour, but historically,
we have reported `EINVAL`, which was the older specified
return value.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
1. change explicit type cast of essential character type, complying with
required [misra-c2012-10.2] rule which states; Expressions of
essentially character type shall not be used inappropriately in addition
and subtraction operations, and
2. add explicit boolean type to 'if' statement controlling expression,
consolidating it with 'buflen' type, thus improving code readability and
maintainability , complying with required [misra-c2012-14.4] rule which
states; ; The controlling expression of an if statement and the
controlling expression of an iteration-statement shall have essentially
boolean type, and
3. add enclosing parentheses enforcing and clarifying precedence of
operators, improving code readability and maintainability, complying
with *advisory* [misra-c2012-12.1] rule which states; The precedence of
operators within expressions should be made explicit.
Found as a coding guideline violation (Rules 10.2, 14.4), and coding
guideline recommendation (Rule 12.1) by static code scanning tool.
Note: Tested on STM32L5 Nucleo-144 board (stm32l552xx).
Signed-off-by: ferar alashkar <ferar.alashkar@gmail.com>
change explicit type cast of essential character type, complying with
required [misra-c2012-10.2] rule which states; Expressions of
essentially character type shall not be used inappropriately in addition
and subtraction operations.
Found as a coding guideline violation (Rule 10.2) by static code
scanning tool.
Note: Tested on STM32L5 Nucleo-144 board (stm32l552xx).
Signed-off-by: ferar alashkar <ferar.alashkar@gmail.com>
add explicit unsigned suffices to various immediate numbers, matching
them to size_t, complying with required [misra-c2012-10.4] rule which
states; Both operands of an operator in which the usual arithmetic
conversions are performed shall have the same essential type category.
Found as a coding guideline violation (Rule 10.4) by static code
scanning tool.
Note: Tested on STM32L5 Nucleo-144 board (stm32l552xx).
Signed-off-by: ferar alashkar <ferar.alashkar@gmail.com>
Add a name to the choice group for selecting the C++
standard to be able to override the default standard in
Kconfig.* files.
Signed-off-by: Damian Krolik <damian.krolik@nordicsemi.no>
This change allows users to call pthread_create() with
the pthread_attr_t argument equal to NULL.
If Zephyr is configured with `CONFIG_DYNAMIC_THREAD`, then a
suitable thread stack will be allocated via
k_thread_stack_alloc(). The allocated thread stack is
automatically freed via k_thread_stack_free().
This makes the Zephyr implementation of pthread_create()
compliant with the normative spec.
Signed-off-by: Christopher Friedt <chrisfriedt@gmail.com>
Since the argument is a 32-bit unsigned int, all possible
values satisfy the condition that intval < UINT64_MAX - 1.
Remove the redundant conditional.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
After commit 9a0aebc5fd,
the exclusion of qemu_x86_tiny is no longer and the "depends on"
option was removed. However, the comment about that remained.
Remove the comment as it is no longer valid.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Skip child objects and arrays that are not specified in the given object
descriptor when parsing a JSON input string.
This patch adds support for extra child arrays which previously were not
supported by the parser as opposed to additional child objects.
Fixes#47988
Signed-off-by: Markus Fuchs <markus.fuchs@ch.sauter-bc.com>
Changed initial guess from a simple x/3 to dividing the exponent by 2.
This makes large or small numbers like 10e10 and 01e-10 converge in a few
loops.
Added a loop counter to ensure that the algorithm breaks out of the loop in
the case that the algorithm doesn't converge (toggling between two
numbers).
Added test cases for sqrt and sqrtf in libc. Tested with a range of numbers
between 10e10 and 10e-10. Verify good accuracy in test case.
Closes: #55962
Signed-off-by: Lawrence King <lawrencek52@gmail.com>
Multiple reader threads unlocking the read lock simultaneously might
cause the program hang because it's possible that no thread is
identified as the last one to active the writer thread.
To fix the issue, swap the k_sem_give sequence.
Signed-off-by: Jaxson Han <jaxson.han@arm.com>
Picolibc has both recursive and non-recursive mutex uses. The bulk of the
library locking uses the global libc lock, which is a recursive mutex as
that greatly simplifies the implementation.
The only use of non-recursive mutexes is in the stdio code when dealing
with file system I/O via fopen.
Using mutexes for both APIs is valid; the assumption picolibc makes is that
the non-recursive mutexes are somehow cheaper or faster and should be
preferred. However, in Zephyr, recursive mutexes are the default and the
non-recursive locks for picolibc were implemented using semaphores.
Switch the non-recursive picolibc locks to just invoking the existing
recursive functions using mutexes. This avoids pulling in another lock
implementation, saving a bit of space.
This also lets the kernel.memory_protection.mem_map test work on
qemu_x86_tiny where the amount of memory available is 320kB and that is
nearly filled by this test case, leaving too little space for allocating
pages in the k_mem_map_unmap test.
Signed-off-by: Keith Packard <keithp@keithp.com>
The `pthread_once_lock` `k_mutex` is statically initialized and
only visible within file scope. Coverity identified it as unsafe
because the return values of `pthread_mutex_lock()` and
`pthread_mutex_unlock()` were unchecked. However, if those
functions were to fail here, it would be indicative that
something far worse has happened.
In any case, we add assertions that these functions
succeed rather than silently ignoring with `(void)`, which
ensures that we have coverage when assertions are enabled,
in test, while removing unneeded code with assertions disable,
in production.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
The `struct k_spinlock` size is zero bytes under certain
circumstances. This is a bit of a problem, because it breaks a
number of assumptions about things in C.
That should be fixed when #59922 is addressed.
This change is just a hotfix to correct the specific condition
where we will need workarounds in other source files.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
This change is setting up for switching over to proper POSIX
option requirements, feature test macros, and a dependency
structure that is reflective of the standard.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Most of the posix source files can be easily identified by a
short name. I.e. most of the `pthread_` prefixed files do not
need the `pthread_` prefix.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
A significant enough portion of these files has been
changed to justify adding Meta copyright as well as
that of the original author.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Now that the gcc compiler driver uses the -T flag instead of -Wl,-T, we can
remove the hack here that kept the picolibc specs file from inserting the
picolibc linker script.
Signed-off-by: Keith Packard <keithp@keithp.com>
It is possible to build with the PICOLIBC_MODULE
with the POSIX arch targets which use the native
simulator as runner.
Update filtering accordingly.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
It is possible to build with STDOUT_CONSOLE with
the embedded C libraries with the POSIX arch.
Narrow down the filtering accordingly.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
So they depend or select on the right NATIVE_BUILD
instead of NATIVE_APPLICATION.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Add ACPI support for Zephyr using acpica open source
project. ACPI subsystem use to discover and configure
hardware components, perform power management (e.g. putting
unused hardware components to sleep), auto configuration (e.g.
Plug and Play and hot swapping) etc.
Signed-off-by: Najumon Ba <najumon.ba@intel.com>
This patch fixes support for encoding and decoding multidimensional arrays
as described by the JSON_OBJ_DESCR_ARRAY_ARRAY() macro.
Currently, the JSON array encoding and decoding functions, arr_encode()
and arr_parse(), expect array elements to be of object or primitive type.
However, arrays may be nested and so an array's elements may also be
arrays.
In order to support nested arrays, two special cases must be considered:
1. The array of objects/arrays sub-descriptor is described by two
`json_obj_descr` structs and so two instead of one `json_obj_descr`
structs must be skipped when iterating over the JSON descriptor to get to
an array's elements.
2. The implicit array item count field has to be considered for the
parent itself and all its child array items when calculating an element's
size.
Fixes#50801
Signed-off-by: Markus Fuchs <markus.fuchs@ch.sauter-bc.com>
Fixes#58911. Previously the stat command returned
information on the filesystem, but not the file itself.
Because block size is still set this function is
backwards compatible with the previous faulty
behavior.
Signed-off-by: Vincent van Beveren <v.van.beveren@nikhef.nl>
The POSIX API compatibility shim can be used for some
of the POSIX ARCH targets.
Narrow the Kconfig filtering accordingly.
Note that the recommended configuration when building
with the native simulator is still to use an embedded
C library. Using the host C library will in some cases
cause undesired behaviour.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Provide a sensible default for the POSIX architecture,
as now it is possible to build with it.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
The spin loop in _exit() needs a Z_SPIN_DELAY() for the
POSIX architecture, so it does not hang the whole
executable on that infinite loop but only the thread
that exit'ed.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
The spin loop in _exit() needs a Z_SPIN_DELAY() for the
posix architecture, so it does not hang the whole
executable on that infinite loop but only the thread
that exit'ed.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Fix a regression introduced by commit e6eb0a705b ("posix: eventfd: revise
locking, signaling, and allocation"), which was a complete rewrite stating
that:
The `wait_q` and `k_poll_signal` entries were removed from
`struct eventfd` as they were unnecessary.
In fact, `k_poll_signal` (both `read_sig` and `write_sig`) were used to
wake-up blocking `poll()` invocation in another thread. This is no longer
the case now, i.e. `poll(..., POLLIN)` does not return after calling
`eventfd_write()` on the observed (polled) FD.
Fix this regression by bringing back `read_sig` and `write_sig` to very
similar state as it was before.
Fixes: e6eb0a705b ("posix: eventfd: revise locking, signaling, and
allocation")
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
This adds a few line use zephyr_syscall_header() to include
headers containing syscall function prototypes.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
portability.posix.common.arcmwdtlib test fails with ARCMWDT libc.
This path fixes the test.
STDIN_FILENO and others macroses are used in libc-hooks.c only. So they
defined localy.
Signed-off-by: Nikolay Agishev <agishev@synopsys.com>
Previously, the `posix_internal.h` header needed to be exposed
to the application because we had non-trivial details for
most posix types (pthread, mutex, cond, ...). Since most of
those have been simplified to a typedef'ed integer, we
no longer need to expose that header to the applicaiton.
Additionally, it means that we can adopt normalized
header order in posix.
Additionally, keep more implementation details hidden
and prefer the static keyword on internal symbols where
possible.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
To enable testing, introduce `CONFIG_PTHREAD_CREATE_BARRIER`.
Some observations were made that running several Qemu SMP targets
concurrently could lead to synchronization problems. On such
targets, it was found that the synchronization issues were
mitigated by introducing a `pthread_barrier_t` shared between
`pthread_create()` and the spawned thread.
It is suggested to enable the option when running many
SMP tests concurrently in several parallel Qemu processes,
e.g. with `twister`.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Previously, pthreads suffered from some race conditions.
This was almost inevitable given that it was maintained in
parallel to Zephyr's threading and synchronization API.
The unfortunate side-effect of with that is that it did not
receive the reliability and other improvements that
`k_thread`s did.
Here, we perform a significant update of pthread code so
that it depends directly on public Zephyr API. With that,
we reuse as many concepts as possible and pthreads benefits for
free from any improvement made to Zephyr's threading and
synchronization APIs.
Included with this change, we
* implement state with `ready_q`, `run_q`, and `done_q`
* use `pthread_barrier_wait()` to sync `pthread_create()`
* synchronize internal state with a spinlock
These pthreads are considerably more reliable than
before.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
The internal representation of `pthread_cond_t`,
`struct posix_cond`, is an identical clone of `struct k_condvar`
but without the benefit of being able to use all of the
existing `k_condvar_*()` suite of functions.
The first step in the right direction was switching
the external representation of `pthread_cond_t` to a simple
`int`. Let's take the next step in the right direction, which
is getting rid of `struct posix_cond`.
For now, let's keep this change as a simple type
substitution. Eventually, we should be able to fully switch
to Zephyr API internally.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
The internal representation of `pthread_mutex_t`,
`struct posix_mutex`, is basically a clone of `struct k_mutex`
but without the benefit of being able to use all of the
existing `k_mutex_*()` suite of functions.
The first step in the right direction was switching
the external representation of `pthread_mutex_t` to a simple
`int`. Let's take the next step in the right direction, which
is getting rid of `struct posix_mutex`.
The only significant difference between `struct k_mutex` and
`struct posix_mutex` is that the latter needs a `type` field.
Since there were a fixed number of `struct posix_mutex`, we
can just externalize the `type` field and reuse
`struct k_mutex` as-is.
For now, let's keep this change as a simple type
substitution. Eventually, we should be able to fully switch
to Zephyr API internally.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Previously pthread_barrier_t was implemented in terms of wait
queues and internal scheduler functions.
This introduced some obstacles and inconsistency. In order
to be more consistent, rely only on Zephyr's public API and
reuse as many concepts as possible.
Deprecate `PTHREAD_BARRIER_DEFINE()` since it's non-standard.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Deprecate `EFD_IN_USE` and `EFD_FLAGS_SET` as they are not
specified as part of any public `eventfd()` API.
While those are being deprecated, use `_INTERNAL` variants.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
TL;DR - a complete rewrite.
Previously, the prototypical `eventfd()` usage (one thread
performing a blocking `read()`, followed by another thread
performing a `write()`) would deadlock Zephyr. This shortcoming
has existed in Zephyr's `eventfd()` implementation from the
start and the suggested workaround was to use `poll()`.
However, that is not sufficient for integrating 3rd-party
libraries that may rely on proper `eventfd()` blocking
operations such as `eventfd_read()` and `eventfd_write()`.
The culprit was the per-fdtable-entry `struct k_mutex`.
Here we perform a minor revision of the locking strategy
and employ `k_condvar_broadcast()` and `k_condvar_wait()`
to signal and wait on the holder of a given `struct k_mutex`.
It is important to note, however, that the primary means of
synchronizing the eventfd state is actually the eventfd
spinlock. The fdtable mutex and condition variable are mainly
used for the purposes of blocking io (r,w,close) and are not
used in the code path of non-blocking reads.
The `wait_q` and `k_poll_signal` entries were removed from
`struct eventfd` as they were unnecessary.
Additionally, switch to using a bitarray because it is
possibly faster than linear search for allocating and
deallocating eventfd resources.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
There is no major benefit to haveing `eventfd_read()` and
`eventfd_write()` as `static inline` functions and it is
arguably much easier to solve the deadlock bug when they
are not `inline`.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Complement the `struct k_mutex` in each fdtable entry
with a `struct k_condvar`. The reasoning for this should be
self-evident.
For a bit of history, `fdtable` was introduced in
commit 06eb489c45 ("kernel: add condition variables")
which predates `struct k_condvar`, introduced in
commit f484bbaa26 ("lib: posix: Implement generic file descriptor table")
by almost 2 years.
Additionally, provide a new accessor function,
`z_get_obj_lock_and_cond()`, that (optionally) gets the mutex
and condition variable associated with the provided object and
vtable.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Coverity does not like that we are passing a pointer to a location
just beyond fixed array. Inside the function access is done through
negative indexes so there was no memory corruption but to satisfy
Coverity pointer to the last element of the array is passed and
we start from index 0 instead of -1.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
People interested in those options may be truly concerned by binary
sizes. Let's provide complete information.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Compiling using clang fails with following error:
error: unannotated fall-through between switch labels
[-Werror,-Wimplicit-fallthrough]
Actually, missing 'break' statement is not an error here because
the next label is 'default' which contains 'break' only. For the same
reason we can add it to silence the compiler.
Signed-off-by: Patryk Duda <pdk@semihalf.com>
This matches the size that would be used with the newlib_nano configuration
and allows several tests to complete which would otherwise fail due to
insufficient heap space.
Signed-off-by: Keith Packard <keithp@keithp.com>
arm, arc and riscv all have special cases for malloc arena alignment that
might be smaller than the minimum required for a C allocator. In
particular, the riscv value might actually be zero, which turns out to be
an invalid alignment value.
Make sure all of these have alignment that meets the C language
requirements for allocation alignment.
Signed-off-by: Keith Packard <keithp@keithp.com>
With Zephyr now always using `int main(void)`, there's no longer any need
for this definition. The last remaining use which gated the declaration of
_posix_zephyr_main isn't necessary as adding that declaration
unconditionally is harmless.
Signed-off-by: Keith Packard <keithp@keithp.com>
We are currently reporting the wrong mismatching bits in in-between
bundles. Fix this and extend the test to cover the wrong case.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
When using the Zephyr SDK toolchain, prefer the pre-built version of
picolibc over using the picolibc module. This will reduce the time it takes
to build applications.
Signed-off-by: Keith Packard <keithp@keithp.com>
Until now iterable sections APIs have been part of the toolchain
(common) headers. They are not strictly related to a toolchain, they
just rely on linker providing support for sections. Most files relied on
indirect includes to access the API, now, it is included as needed.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Analog to json_obj_encode vs. json_calc_encoded_len which
calculates the object len using json_obj_encode, introduce
json_calc_encoded_arr_len which calculates the length using
json_arr_encode. That is needed when the object to be encoded
is array on the root level.
Signed-off-by: Miika Karanki <miika.karanki@vaisala.com>
For the POSIX architecture we do not define an
empty __cxa_atexit() as it otherwise would
override its host libC counterpart. And this would
both disable the atexit() hooks, and prevent possible
test code global destructors from being registered.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
BSD-3-Clause-UC is not a valid license name and spdx validator
generates a Warning! when detected. Most probably a typo from
old version of strtoll.c licenced as BSD-4-Clause-UC.
see commit: 570ed08
Signed-off-by: Lucian Zala <zala.lucian@gmail.com>
The malloc arena needs to be aligned to a suitable size on Risc-V, and the
Z_RISCV_STACK_GUARD_SIZE variable is a helpful proxy for what size that
is. However, that is not always a power of two, so round it up to make the
linker capable of performing the necessary allocation.
Signed-off-by: Keith Packard <keithp@keithp.com>
This function may be used by system libraries (like libstdc++), so instead
of gating compile on target language, always build this function and gate
API visibility in header files instead.
Signed-off-by: Keith Packard <keithp@keithp.com>
Not all xtensa targets define the top of usable RAM via the _heap_sentry
address; it looks like the list is limited to esp32, esp32s2, esp32s3 and
intel parts. The first three all define HAS_ESPRESSIF_HAL, so key the test
using that or SOC_FAMILY_INTEL_ADSP.
Signed-off-by: Keith Packard <keithp@keithp.com>
Instead of explicitly initializing the mutex at runtime, use
SYS_MUTEX_DEFINE to initialize it at build time instead.
Signed-off-by: Keith Packard <keithp@keithp.com>
Have eventfd Kconfig select POLL is the code utilizes the polling
API. We get a link error for tests/lib/fdtable/libraries.os.fdtable
when building on arm-clang without this.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
Enable the common malloc implementation when using picolibc. Support
existing Picolibc configurations by respecting existing PICOLIBC_HEAP_SIZE
settings.
When PICOLIBC_HEAP_SIZE is set to a value other than -2, then
*always* set COMMON_LIBC_MALLOC_ARENA_SIZE to that value.
Signed-off-by: Keith Packard <keithp@keithp.com>
When using the common malloc implementation on systems not using the
minimal C library, allocate a reasonable default malloc heap according to
the following rules (adopted from the Picolibc heap size defaults):
* When an MMU is available, allocate 16kB.
* When USERSPACE is enabled for a device with an MPU require PoT alignment,
allocate 1024 bytes
* Otherwise, use all available memory.
Signed-off-by: Keith Packard <keithp@keithp.com>
Change the default value of MINIMAL_LIBC_MALLOC_ARENA_SIZE from 0 to an
otherwise invalid value of -2. This allows the common malloc code to
distinguish between an application which explicitly sets the malloc arena
to zero bytes from an application which does not set the variable at all.
When MINIMAL_LIBC_MALLOC_ARENA_SIZE is set to a value other than -2, then
*always* set COMMON_LIBC_MALLOC_ARENA_SIZE to that value. Emit a compile
warning to let the developer know that their application is using a
deprecated Kconfig variable.
This means that MINIMAL_LIBC_MALLOC_ARENA_SIZE now always overrides
COMMON_LIBC_MALLOC_ARENA_SIZE, but an application setting both will already
be getting a configuration warning for setting the DEPRECATED value
anyways, which should serve as an adequate indication to the user.
Signed-off-by: Keith Packard <keithp@keithp.com>
When CONFIG_COMMON_LIBC_MALLOC_ARENA_SIZE is set to -1, make the malloc
arena use all remaining RAM. When an MMU exists, allocate the arena at
startup using k_mem_map. Otherwise, compute the available memory
automatically and use that.
When an MPU is being used to manage the malloc arena, make sure the heap
respects any MPU alignment requirements. Otherwise, align the heap to
sizeof(double).
Signed-off-by: Keith Packard <keithp@keithp.com>
Move the abort implementation into common so its shared among the
libc. As part of this start using the common abort on newlib.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
Move the strnlen implementation into common so its available to any
libc that may not implement strnlen.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
Introduce Kconfig symbol FULL_LIBCPP_SUPPORTED that is similar to the
C version FULL_LIBC_SUPPORTED. This way we can utilize the same
pattern in tests and samples to filter for when a full libc++ is
needed.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
The `SCHED_OTHER` scheduling priority is mandatory as part of
POSIX. It must be numerically distinct from `SCHED_FIFO`,
`SCHED_RR`, and `SCHED_SPORADIC`, but is implementation-
defined and may behave identically to `SCHED_FIFO` or
`SCHED_RR`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Select FULL_LIBC_SUPPORTED when picolibc is available.
Add picolibc as a secondary default C library when REQUIRES_FULL_LIBC is
selected. This is necessary as tests gated on FULL_LIBC_SUPPORTED need to
be sure that a full C library will be selected -- if only picolibc is
available, those tests will need to select that.
This should permit use of a picolibc-only crosstool-ng toolchain in
testing.
Signed-off-by: Keith Packard <keithp@keithp.com>
This symbol is selected when the target has any full libc available. This
allows tests to filter on this condition. It doesn't depend on whether the
application actually selects that library, only whether requesting a full C
library would work.
Signed-off-by: Keith Packard <keithp@keithp.com>
This reflects whether newlib is available in the environment. This
symbol should be used in place of TOOLCHAIN_HAS_NEWLIB.
Signed-off-by: Keith Packard <keithp@keithp.com>
Add a prompt to the Kconfig symbol so that applications can select this in
their configuration to guide C library selection away from the minimal C
library.
Signed-off-by: Keith Packard <keithp@keithp.com>
This will (eventually) permit use of a common malloc implementation with
other C libraries, reducing the amount of Zephyr-specific code required
to support each C library.
Signed-off-by: Keith Packard <keithp@keithp.com>
This allows the C library common library to be empty if no pieces are
needed, eliminating cmake warnings.
Signed-off-by: Keith Packard <keithp@keithp.com>
Add a prompt to the Kconfig symbol so that applications can select this in
to guide C++ standard library selection towards configurations which
provide a complete implementation.
Signed-off-by: Keith Packard <keithp@keithp.com>
Add limits.h header that defines PATH_MAX as this is not defined
by the toolchain headers.
The value of 256 for PATH_MAX is based on how it is set in
lib/libc/minimal/include/limits.h.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
This ensures that the compiler won't end up recognizing patterns in code
that map to known built-in functions, like noticing that the memcpy
implementation looks a lot like a memory copy operation that could be
replaced by a call to ... memcpy.
When built with -ffreestanding, this has no effect as that option always
enables -fno-builtin.
Signed-off-by: Keith Packard <keithp@keithp.com>
When gcc is building without -fno-builtin, it will optimize calls like
printf("\n") into a call to putchar('\n'), but it won't use a static inline
in that case, instead insisting on a real function.
To make this a bit easier, adopt the usual C library practice of making
putc and putchar macros instead of static inline functions. There's no loss
of typechecking as the parameters are directly passed to underlying
functions with the same parameter types.
Signed-off-by: Keith Packard <keithp@keithp.com>
Remove statement probably left after rebase. ret should be 0 or
error codes, described in docs, and it is rewritten 4 lines below.
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
In d540cf8877 I tried to optionally enable the cache management
functions in Open-AMP introducing a new CONFIG_OPENAMP_WITH_DCACHE
symbol.
This is not working. Introduce a proper fix to have this actually
working correctly as intended.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
When using the toolchain C library, that must be added to the link command
after all other libraries and modules in the system to resolve undefined
symbols.
Signed-off-by: Keith Packard <keithp@keithp.com>
Add an explicit compiler configuration, COMPILER_FREESTANDING, which
controls whether the compiler should operate in freestanding or hosted mode
(according to the C and C++ language specifications.
This depends on having a C library which conforms with the language
specification, and the minimal C library does not. Have the minimal C
library select COMPILER_FREESTANDING to continue using freestanding mode
with that library.
For other C libraries, leave this disabled by default while allowing users
to enable it if they want to go back to the previous configuration.
Signed-off-by: Keith Packard <keithp@keithp.com>
MISRA Rule 5.7 requires uniqueness of tag identifiers. Shell is
frequently problematic because many code uses `const struct shell
*shell`. This causes CI noise every time one of these shell files is
edited, so let's update all of them with `const struct shell *sh`
instead.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The init infrastructure, found in `init.h`, is currently used by:
- `SYS_INIT`: to call functions before `main`
- `DEVICE_*`: to initialize devices
They are all sorted according to an initialization level + a priority.
`SYS_INIT` calls are really orthogonal to devices, however, the required
function signature requires a `const struct device *dev` as a first
argument. The only reason for that is because the same init machinery is
used by devices, so we have something like:
```c
struct init_entry {
int (*init)(const struct device *dev);
/* only set by DEVICE_*, otherwise NULL */
const struct device *dev;
}
```
As a result, we end up with such weird/ugly pattern:
```c
static int my_init(const struct device *dev)
{
/* always NULL! add ARG_UNUSED to avoid compiler warning */
ARG_UNUSED(dev);
...
}
```
This is really a result of poor internals isolation. This patch proposes
a to make init entries more flexible so that they can accept sytem
initialization calls like this:
```c
static int my_init(void)
{
...
}
```
This is achieved using a union:
```c
union init_function {
/* for SYS_INIT, used when init_entry.dev == NULL */
int (*sys)(void);
/* for DEVICE*, used when init_entry.dev != NULL */
int (*dev)(const struct device *dev);
};
struct init_entry {
/* stores init function (either for SYS_INIT or DEVICE*)
union init_function init_fn;
/* stores device pointer for DEVICE*, NULL for SYS_INIT. Allows
* to know which union entry to call.
*/
const struct device *dev;
}
```
This solution **does not increase ROM usage**, and allows to offer clean
public APIs for both SYS_INIT and DEVICE*. Note that however, init
machinery keeps a coupling with devices.
**NOTE**: This is a breaking change! All `SYS_INIT` functions will need
to be converted to the new signature. See the script offered in the
following commit.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
init: convert SYS_INIT functions to the new signature
Conversion scripted using scripts/utils/migrate_sys_init.py.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
manifest: update projects for SYS_INIT changes
Update modules with updated SYS_INIT calls:
- hal_ti
- lvgl
- sof
- TraceRecorderSource
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
tests: devicetree: devices: adjust test
Adjust test according to the recently introduced SYS_INIT
infrastructure.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
tests: kernel: threads: adjust SYS_INIT call
Adjust to the new signature: int (*init_fn)(void);
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Add string.h header that declerates prototype for strnlen()
as this is not provided by the toolchain headers.
Add strings.h to allow anything that expects strncasecmp() to
be defined there to build correctly. The arm toolchain actually
defines strncasecmp() in <string.h> so we just have <strings.h>
include <string.h>.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
* Move 'struct _timespec' into sys/_timespec.h as expected by
<posix/time.h>.
* Introduce 'struct timeval' in sys/_timeval.h as expected by
<posix/time.h>.
* Add mode_t to <sys/types.h> as expected by <posix/stat.h>
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
Various headers include <sys/cdefs.h> like zephyr/posix/sys/stat.h.
When building with the arm-clang toolchain we utilize armstdc
so we need an empty cdefs.h similar to what minimal libc has so
things will build.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
We get compile warnings of the form:
error: converting the result of
'<<' to a boolean; did you mean
'((__aeabi_ctype_table_ + 1)[(byte)] << 28) != 0'?
[-Werror,-Wint-in-bool-context]
if (!isprint(byte)) {
^
Since isprint (and the other is* functions) return an int, change check
to an explicit test against the return value.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
We get compile warnings of the form:
error: converting the result of
'<<' to a boolean; did you mean
'((__aeabi_ctype_table_ + 1)[(byte)] << 28) != 0'?
[-Werror,-Wint-in-bool-context]
if (!isprint(byte)) {
^
Since isprint (and the other is* functions) return an int, change check
to an explicit test against the return value.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
Introduce a place to share implementations of libc functions that
are needed by different libc versions. Place time() in this common
location so it can be shared when building for either minimal libc or
armclang libc.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
* add opening and closing braces around several blocks
* convert dos line endings to unix
* move assignments out of conditionals
* ensure adequate spacing after control blocks
Signed-off-by: Chris Friedt <cfriedt@meta.com>
* Add a flexible Hashmap API
* Add a Separate-Chaining Hashmap Implementation
* Add a Open-Addressing Linear Probe Hashmap Implementation
* Add a C-Wrapper for `std::unordered_map` for benchmarking
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Changed incrementing `for` loop counters to `size_t` from `int` to
eliminate warning, "warning: comparison of integer expressions of
different signedness: 'uint32_t' {aka 'unsigned int'} and 'int'
[-Wsign-compare]"
Signed-off-by: Zachary J. Fields <zachary_fields@yahoo.com>
Changed incrementing `for` loop counters to `size_t` from `int` to
eliminate warning, "warning: comparison of integer expressions of
different signedness: 'uint32_t' {aka 'unsigned int'} and 'int'
[-Wsign-compare]"
Signed-off-by: Zachary J. Fields <zachary_fields@yahoo.com>
Disable tests/kernel/mem_protect/syscalls for qemu_arc_em where
we trigger ARC QEMU bug which cause illegal instruction exception
on perfectly valid ARC code.
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Signed-off-by: Evgeniy Paltsev <PaltsevEvgeniy@gmail.com>
This is a workaround for a compiler bug on (at least) GCC 12.1.0 in
Zephyr SDK 0.15.1. The optimizer generates this function with a last
instruction that is an unconditional branch (a tail call into the
chunk_set() handling). But that means that the NEXT instruction gets
decoded as part of the branch delay slot, but that instruction isn't
part of this function! Some instructions aren't legal in branch delay
slots. One of those is ENTER_S, which is a very common entry
instruction for whatever function the linker places after us. It
seems like the compiler doesn't understand this problem. Stuff a NOP
in to guarantee the code is legal.
Comment above is duplicated in the code. The workaround is
straightforward once the issue is understood, but the path to get here
was hilariously weird.
Fixes#54720
Signed-off-by: Andy Ross <andyross@google.com>
Clang support still work in progress in zephyr. So for now enable only
on gcc based toolchains.
We should enable this back for more toolchains once we have full clang
support and are able to verify.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
With all of the test suite issues on this board fixed, qemu_x86_tiny should
now be usable with picolibc.
Signed-off-by: Keith Packard <keithp@keithp.com>
Picolibc now supports all of the Zephyr SDK target architectures for C.
qemu_x86_tiny needs fixes to get the libc partition linked
into the right spot. See issue #54148
Signed-off-by: Keith Packard <keithp@keithp.com>
The limiting factor is the output bitmask that says which elements have
been filled in by the parser. This patch changes the bitmask type from int
to int64_t.
Signed-off-by: Björn Stenberg <bjorn@haxx.se>
Although the eventfd API is not (yet) a part of POSIX,
it's pretty well ubiquitous on POSIX systems now.
Enable it by default when `CONFIG_POSIX_API=y`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
The `getopt()` function is part of POSIX and should be
available when applications choose to enable general POSIX
API support.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
With the `<zephyr/posix/...> prefix, it became
exponentially more difficult to integrate 3rd-party
libraries that depend on the POSIX API.
Standard POSIX headers should be available in standard
include paths - and that should most certainly the case
when `CONFIG_POSIX_API=y`.
With this change:
* When `CONFIG_POSIX_API=y`
- applications have explicitly chosen to use
POSIX APIs.
- all standard POSIX includes are in the default
include path.
* When `CONFIG_POSIX_API=n`
- applications *may* include POSIX headers
explicitly with the namespaced prefix
- e.g. `#include <zephyr/posix/unistd.h>`
- individual Kconfig options can be used to
enable POSIX features selectively, such as
`getopt` or `eventfd`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Refactor sys_clock_disable not implemented behavior.
This follows the coding guidelines
Rule A.1: Conditional Compilation:
Do not conditionally compile function declarations in header files.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
libstdc++ is supported with Picolibc only when the toolchain version of
Picolibc is use -- libstdc++ must be built using a specific Picolibc build
and libstdc++ is included with the toolchain.
Ideally, we'd allow the use of the Picolibc module whenever we weren't
using the GNU libstdc++, including when using the minimal libc++. However,
the obvious dependency settings create a loop:
config PICOLIBC
depends on PICOLIBC_SUPPORTED
config PICOLIBC_SUPPORTED
depends on !(GLIBCXX_LIBCPP && "$(ZEPHYR_TOOCHAIN_VARIANT" = "zephyr")
config GLIBCXX_LIBCPP
depends on NEWLIB_LIBC || PICOLIBC
To break this loop, we replace GLIBCXX_LIBCPP in the second block with
CPP:
config PICOLIBC_SUPPORTED
depends on !(CPP && "$(ZEPHYR_TOOCHAIN_VARIANT" = "zephyr")
This means that picolibc cannot be used with any C++ apps when using the
Zephyr SDK, even when not using the GNU libstdc++. However, Zephyr SDK 0.16
will come with an additional Kconfig file that includes:
config PICOLIBC_SUPPORTED
def_bool y
depends on "$(ZEPHYR_TOOLCHAIN_VARIANT)" = "zephyr"
This will override the Kconfig bits included in Zephyr and allow use of the
Picolibc module with C++ code, including using the minimal libc++ bits.
Signed-off-by: Keith Packard <keithp@keithp.com>
This ensures that all pending writes are committed. This is relevant for
flags in .noinit SRAM that are read back after reset.
Signed-off-by: Armin Brauns <armin.brauns@embedded-solutions.at>
This commit reworks the C++ standard library configurations such that:
* the separation between the Zephyr minimal C++ library and the fully
featured C++ standard libraries, such as GNU libstdc++, is clear.
This is done by deprecating the Kconfig `CONFIG_LIB_CPLUSPLUS`
symbol, which implies that the minimal C++ library is selected when
set to `n`, and introducing the `CONFIG_MINIMAL_LIBCPP` symbol.
* the type of the selected C++ standard library is clear. This is done
by introducing a Kconfig choice, `LIBCPP_IMPLEMENTATION`, for the C++
standard library type and providing the choice symbols for each
library type supported, such as `CONFIG_MINIMAL_LIBCPP` and
`CONFIG_GLIBCXX_LIBCPP`.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
The Kconfig `CONFIG_RTTI` symbol enables the C++ Run-Time Type
Information (RTTI) support and is a feature of the standard C++
library; therefore, it should be prefixed with `CPP_` as with the rest
of the C++-specific symbols.
This commit renames the Kconfig `CONFIG_RTTI` symbol to
`CONFIG_CPP_RTTI` in order to better align with the other existing C++
Kconfig symbols, and deprecates the old `CONFIG_RTTI` symbol.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
`CONFIG_EXCEPTIONS` is ambiguous in that the word "exception" may be
used to refer to many different types of exceptions, from the hardware
generated faults and interrupts (aka. exceptions) to the C++
exceptions.
This commit renames the Kconfig `CONFIG_EXCEPTIONS` symbol to
`CONFIG_CPP_EXCEPTIONS` in order to clarify that the symbol refers to
the C++ exceptions and eliminate any ambiguity, and deprecates the old
`CONFIG_EXCEPTIONS` symbol.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
This commit updates all in-tree code to use `CONFIG_CPP` instead of
`CONFIG_CPLUSPLUS`, which is now deprecated.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
There is no reason for the top-level C++ Kconfig symbol to be called
`CPLUSPLUS` when the convention used throughout the Zephyr codebase is
to use "cpp" to refer to C++ and the rest of the C++-related Kconfig
symbols are prefixed with `CPP_`.
This commit renames the Kconfig `CPLUSPLUS` symbol to `CPP` in order to
better align with the de-facto standard naming convention used throughout
the Zephyr codebase, and deprecates the old `CPLUSPLUS` symbol.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
This commit relocates the "C++ minimal library" components, that
implement a very limited subset of the standard C++ library, to a
dedicated directory, `lib/cpp/minimal`, in order to provide a clear
separation among the different C++ library components.
After this refactoring, the top `lib/cpp` directory should only contain
the sub-directories for each C++ library (i.e. `abi`, `minimal`). In
the future, a C++ library-specific shim layer implementation may be
added as sub-directories under `lib/cpp` as well (e.g. `libstdc++`
sub-directory containing the shim layer implementation for the GCC
libstdc++ library) -- this is similar to how the libc directories are
structured under `lib/libc`.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
This commit relocates the "C++ ABI library" components such as
global constructor/destructor and initialiser handlers to a dedicated
directory, `lib/cpp/abi`, in order to provide a clear separation
between the C++ ABI/runtime library and the standard C++ library
components.
Note that the Zephyr C++ ABI library currently implements the GNU/GCC
C++ ABI, which is the de-facto standard ABI used by many compilers
including Clang -- it may be necessary to sub-divide the `lib/cpp/abi`
into `lib/cpp/abi/gnu` and `lib/cpp/abi/someotherabi` in the future
when adding the support for a C++ compiler that expects an ABI vastly
different from the GNU C++ ABI.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
This commit moves the files under `subsys/cpp` directory to the
`lib/cpp` directory because the C++ ABI runtime library and the
standard C++ library components are not a "subsystem" (aka. API) in
conventional sense and is better described as a "library."
Classifying the C++ ABI runtime library and the standard C++ library as
"libraries" instead of "subsystems" also better aligns with how the
existing C standard library implementation (`lib/libc`) is handled.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
After fcntl.h moved to posix, there have a compiler note
on fdtable.c. As suggested in fcntl.h, instead with
zephyr/posix/fcntl.h.
Signed-off-by: HaiLong Yang <hailong.yang@brainco.cn>
The remaining types that needed to be harmonized between
Newlib and Zephyr's POSIX definitions are:
* `struct sched_param`
- don't re-define if using minimal libc
* `pthread_attr_t`
- convert to `struct pthread_attr`
- define type if using minimal libc
- assert acceptible object size
* `pthread_mutexattr_t`
- convert to `struct pthread_mutexattr`
- define type if using minimal libc
- assert acceptible object size
* `pthred_condattr_t`
- convert to `struct pthread_condattr`
- define type if using minimal libc
- assert acceptible object size
* `pthread_once_t`
- adopt newlib definition
- define type if using minimal libc
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Define `PTHREAD_CREATE_DETACHED` and
`PTHREAD_CREATE_JOINABLE` to be compatible with the Newlib
definitions.
This is a temporary workaround for #51211 until Newlib
headers are pulled in.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
The `fcntl.h` header has never been a part of ISO C so move it to
`include/zephyr/posix`.
To ensure a smooth migration, a header was left in
`lib/libc/minimal/include` that prints a deprecation warning.
Users should either include `<zephyr/posix/fcntl.h>` or switch to
`CONFIG_POSIX_API=y`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
The `sys/stat.h` header has never been a part of ISO C so move it
to `zephyr/include/posix/sys/`.
To ensure a smooth migration, leave a stub header in
`lib/libc/minimal/include/sys/` that prints a deprecation warning
suggesting developers either include `<zephyr/posix/sys/stat.h>`
or use `CONFIG_POSIX_API=y`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
The previous method used to calculate seconds in `clock_gettime()`
seemed to have an inaccuracy that grew with time causing the
seconds to be off by an order of magnitude when ticks would roll
over.
This change fixes the method used to calculate seconds.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Fixed issues which were leading to failures when producing
and consuming is preempted at various stages.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Use flag instead of word in the buffer. Using this method allows
to dedicate full buffer capacity for data.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
should be able to configure the time spent waiting
for available resources when calling timer_create()
to not cause a hiccup in applications that require
faster response times than the original hard-coded 100 ms.
Signed-off-by: Nicholas Lowell <nlowell@lexmark.com>
Currently OpenAMP is unconditionally compiled with the options to use
the vrings cache operations disabled.
Add a new CONFIG_OPENAMP_WITH_DCACHE Kconfig option to enable the
support for d-cache operations in OpenAMP when needed.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
This commit updates the Picolibc configurations to remove any
unnecessary defaults and dependencies and conform to the de-facto
standard convention across the Zephyr repository.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
This commit relocates the newlib libc configurations under the top-
level libc Kconfig to a new Kconfig under `lib/libc/newlib` for
improved organisation of the libc configurations.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
This commit relocates the minimal libc configurations under the top-
level libc Kconfig to a new Kconfig under `lib/libc/minimal` for
improved organisation of the libc configurations.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
This commit cleans up the top `CMakeLists.txt` for the libc directory
to use `add_subdirectory_ifdef` in order to make the code more
readable.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
This commit properly writes back cached read pointer in spsc buffer
when freeing the memory.
Signed-off-by: Radoslaw Koppel <radoslaw.koppel@nordicsemi.no>
This commit updates the Newlib integration to define `_ANSI_SOURCE`
in order to prevent Newlib from defining POSIX primitives in its
headers when GNU dialect is used (`-std=gnu*`).
Newlib `features.h` defines `_DEFAULT_SOURCE` when `__STRICT_ANSI__`
is not defined by GCC (i.e. when `-std=gnu*`), which results in the
Newlib headers defining POSIX primitives that are in conflict with the
POSIX primitives defined by Zephyr.
Newlib must not define POSIX primitives unless the feature test macros
such as `_POSIX_SOURCE`, `_GNU_SOURCE` and `_DEFAULT_SOURCE` are
explicitly defined.
Note that `-std=gnu` does not imply `_GNU_SOURCE` or `_DEFAULT_SOURCE`
because it is only supposed to instruct the compiler to use the GNU C
language dialect (i.e. GNU C language extensions).
Refer to the GitHub issue #52739 for more details.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
Declarations for `getopt()` should be in `<unistd.h>`
according to the spec. The extended versions `getopt_long()`
and `getopt_long_only()` are declared in `<getopt.h>`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
This seems to have caused build failures in spite of CI being
green in PR 52653.
This reverts commit fc92eb1b37.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Declarations for `getopt()` should be in `<unistd.h>`
according to the spec. The extended versions `getopt_long()`
and `getopt_long_only()` are declared in `<getopt.h>`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
There are several Zephyr shell commands that are used for
transferring data over a possibly unreliable connection such as a
UART into either memory or flash. For example, `flash load` and
`devmem load`.
Make the CRC functions available so that they can be used to
verify the integrity of data transferred over possibly
unreliable connections.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
The original implementation of `usleep()` was not compliant
to the POSIX spec in 3 ways.
- calling thread may not be suspended (because `k_busy_wait()`
was previously used for short durations)
- if `usecs` > 1000000, previously we did not return -1 or set
`errno` to `EINVAL`
- if interrupted, previously we did not return -1 or set
`errno` to `EINTR`
This change addresses those issues to make `usleep()` more
POSIX-compliant.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
In the case that `sleep()` is interrupted, the POSIX spec requires
it to return the number of "unslept" seconds (i.e. the number of
seconds requested minus the number of seconds actually slept).
Since `k_sleep()` already returns the amount of "unslept" time
in ms, we can simply use that.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Consistent with the change of `pthread_t`, `pthread_mutex_t`,
and `pthread_cond_t` to `uint32_t`, we can now also abstract
`pthread_key_t` as `uint32_t` and separate the implementation
detail, hidden from POSIX API consumers.
This change introduces `CONFIG_MAX_PTHREAD_KEY_COUNT`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
None of the operations that `pthread_key_sem` protected were
blocking, so simply make it a spinlock.
Also made the lock static.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Add CONFIG_CRC for building CRC related routines.
CRC routines are now being built for each application, whether used or
not and are add in the build system unconditionally.
Keep CONFIG_CRC enabled by default for now and until all users have
converted to use the new option.
Partial fix for #50654
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Consistent with the change of `pthread_t` from
`struct posix_thread` to `uint32_t`, we can now also abstract
`pthread_cond_t` as `uint32_t` and separate `struct posix_cond`
as an implementation detail, hidden from POSIX API consumers.
This change deprecates `PTHREAD_COND_DEFINE()` in favour of the
(standardized) `PTHREAD_COND_INITIALIZER`.
This change introduces `CONFIG_MAX_PTHREAD_COND_COUNT`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Previously, `pthread_cond_init()` could not actually fail, and
destroying condition variables was a no-op, and it was missing
in `pthread_exit()`.
However, with the change of `pthread_cond_t` to `uint32_t`, and
since those are embedded inside of `struct posix_thread` for the
time being, the pthread code needs to keep track that it is
relinquishes used condition variables when a thread completes.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Previously `PTHREAD_MUTEX_MASK_INIT` was used to mark a
`pthread_mutex_t` as initialized.
The same needs to be done for `pthread_cond_t` and likely others.
Rather than copy-pasting that and a number of inlines that
duplicate the same functionality, simply make it more generic.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Consistent with the change of `pthread_t` from
`struct posix_thread` to `uint32_t`, we can now also abstract
`pthread_mutex_t` as `uint32_t` and separate `struct posix_mutex`
as an implementation detail, hidden from POSIX API consumers.
This change deprecates `PTHREAD_MUTEX_DEFINE()` in favour of the
(standardized) `PTHREAD_MUTEX_INITIALIZER`.
This change introduces `CONFIG_MAX_PTHREAD_MUTEX_COUNT`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Previously, `pthread_mutex_init()` could not actually fail, and
destroying mutexes was a no-op, so it was missing in a couple of
places.
However, with the change of `pthread_mutex_t` to `uint32_t`, and
since those are embedded inside of `struct posix_thread` for the
time being, the pthread code needs to keep track that it is
relinquishes used mutex resources when a thread completes.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
The `pthread_create()` function is not a cancellation point and
iterating over / mutating `posix_thread_pool` is not a blocking
operation, so use a spinlock for the internal `pthread_pool_lock`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
The `struct pthread` and `enum pthread_state` are actually
implementation details specific to Zephyr.
Let's limit the scope where that level of detail is visible.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
In the interest of reducing any layering concerns,
avoid using POSIX locking primitives where necessary.
Note: it is not safe to use a spinlock here, as the
callback function to `pthread_once()` may itself be
a cancellation point.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
While reviewing the dependency between Picolibc POSIX APIs and Zephyr, I
found that the picolibc libc-hooks code copied a bunch of functions from
the newlib version which weren't needed. This required replacing a few
calls to the (now removed) '_write' hook with printk instead.
Signed-off-by: Keith Packard <keithp@keithp.com>
Several other widely-used pthread implementations
abstract `pthread_t` as `uint32_t`. The benefit
there is that we avoid passing around a pointer to
an internal structure (implementation detail).
Additionally, this removes the alias from `k_tid_t`
to `pthread_t` inside of `struct pthread_mutex`.
Signed-off-by: Chris Friedt <cfriedt@meta.com>
Move runtime checks to use arch_num_cpus(). This is to allow
runtime determination of the number of CPUs in the future.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
This adds the necessary modifier to the stdin/stdout/stderr
variables in picolib, and putting into the z_libc_partition.
This allows userspace applications to utilize these variables
for console I/O.
Fixes#51343
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Compliance check complains about static not being the first
modifier. So move them so there are no more complains
in the future.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The newlib nano variant is currently enabled by default when
`CONFIG_NEWLIB_LIBC=y` and the selected toolchain-architecture
combination includes the newlib nano variant support, even if
`CONFIG_NEWLIB_LIBC_NANO` is not selected by the user.
When `CONFIG_NEWLIB_LIBC=y`, this results in the newlib nano variant
being selected for some architectures (e.g. ARC, ARM and RISC-V), while
the full variant is selected for the rest of the architectures.
The above behaviour is problematic because there exist functional
differences between the newlib full and nano variants (e.g. C99 format
modifiers such as `hh`, `ll`, `z`, `j` and `t` are not available in the
newlib nano variant), and this effectively leads to different level of
C standard support across different architectures when
`CONFIG_NEWLIB_LIBC=y`.
This commit fixes this problem by making the `CONFIG_NEWLIB_LIBC_NANO`
not `default y` and requiring its user to explicitly set this symbol to
`y` when they want to use the newlib nano variant.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
Add a new Kconfig and build this code conditionally, so we do not end up
with this file being built for each zephyr app.
Partial fix for #50654
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Extend capabilities of a minimal libc to support C11 capability
to allocate memory with requested alignment.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
The picolibc heap size configuration (`CONFIG_PICOLIBC_HEAP_SIZE`) is
used to set the statically allocated malloc heap size when userspace is
enabled.
The current default heap size of 1048576 bytes (MMU) and 65536 bytes
(MPU) is too large for most platforms that Zephyr supports and may
result in the picolibc tests being filtered out due to the increased
memory footprint of the compiled image (i.e. SRAM overflow).
This commit updates the default picolibc heap size to a more reasonable
16384 bytes for MMU platforms and 1024 bytes for MPU platforms.
Signed-off-by: Stephanos Ioannidis <stephanos.ioannidis@nordicsemi.no>
Add build assert to make sure _RETARGETABLE_LOCKING is enabled in
toolchain, When _RETARGETABLE_LOCKING is enabled, "_LOCK_T" is "__lock"
pointer type, otherwise "_LOCK_T" is "int" type, so there will be the
following compile warnings when toolchain doesn't enable
_RETARGETABLE_LOCKING:
zephyr/lib/libc/newlib/libc-hooks.c:416:13: warning: cast to pointer from
integer of different size [-Wint-to-pointer-cast]
416 | k_sem_take((struct k_sem *)lock, K_FOREVER);
| ^
zephyr/lib/libc/newlib/libc-hooks.c: In function '__retarget_lock_acquire
_recursive':
zephyr/lib/libc/newlib/libc-hooks.c:423:15: warning: cast to pointer from
integer of different size [-Wint-to-pointer-cast]
423 | k_mutex_lock((struct k_mutex *)lock, K_FOREVER);
| ^
...
Signed-off-by: Jiafei Pan <Jiafei.Pan@nxp.com>
When CONFIG_POSIX_CLOCK is enabled, we should have implementation
of gettimeofday() and therefore time(NULL) should return correct
time, instead of -1.
Signed-off-by: Seppo Takalo <seppo.takalo@nordicsemi.no>
