The deprecated CMake module allows Zephyr to move deprecated code to a
centralized location from where it's both easy to get an overview of
deprecated code, and remove the code when it has been deprecated long
enough.
It further improve existing code, as the real CMake code now can be
changed from n-lines of deprecated code into a single line of the form:
find_package(Deprecated COMPONENTS <deprecated_name>)
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Follow-up: #41301
This commit is a rework and cleanup of the target toolchain handling in
Zephyr CMake build system.
Instead of directly loading code a CMake modules for target toolchain
lookup, the target toolchain now follows the CMake `find_package()`
pattern for finding the target toolchain.
This makes it more clear which modules are responsible for finding tools
and which modules provides build integration / features.
The following tools can now be found using `find_package()`:
- Target toolchain: find_package(TargetTools)
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Follow-up: #41301
This commit is a rework and cleanup of the tools handling in Zephyr
CMake build system.
Instead of directly loading code a CMake modules for tool lookup, the
host tools now follows the CMake `find_package()` pattern for finding
programs / tools in module mode.
This makes it more clear which modules are responsible for finding tools
and which modules provides build integration / features.
The following tools can now be found using `find_package()`:
- Zephyr-sdk : find_package(Zephyr-sdk <version>)
- Generic host tools: find_package(HostTools)
This further allows us to decouple the `verify-toolchain` CMake script
part required by `twister` into a tool lookup module and a dedicated
CMake script which utilizes the lookup module.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Removing Zephyr-SDK code for searching and printing messages when Zephyr
SDK is not found in the system.
After Zephyr SDK has been made available on all platforms, including
MacOS and Windows, then `find_package(Zephyr-sdk)` is always invoked
and thus either `find_package(Zephyr-sdk)` will fail or
ZEPHYR_TOOLCHAIN_VARIANT will be defined.
Meaning the removed code block in this commit can no longer be reached.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Cleanup CMake code by moving devicetree compiler detection to dedicated
FindDtc.cmake module.
This allows us to re-use existing CMake find_package() functionality in
module mode and thus reduce CMake logic.
It further has the benefit of being able to directly specify dtc version
like this:
> find_package(Dtc 1.4.6)
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
This moves the test of compile flags set in environment from generic
toolchain to top-level ${ZEPHYR_BASE}/CMakeLists.txt.
In order to cleanup and make Zephyr CMake modules more generic then
the testing of specific compile flags and warn user is not really a
toolchain module responsible but a implementation detail of the project
itself as to how compile flags should be applied in a project.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
This commit sets the minimum required Zephyr SDK version to 0.15, which
is required to build the current Zephyr codebase due to the following
breaking changes:
* RISC-V ISA spec change to 20191213 requiring the Zicsr and Zifencei
extensions to be specified to the `-march` flag
* ARC64 output format name change to `elf64-littlearc64`
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
This PR allows the user to add symbols to the nocache
section. The use for this could be as follows
zephyr_linker_sources_ifdef(CONFIG_NOCACHE_MEMORY
NOCACHE_SECTION
nocache.ld
)
nocache.ld (as shown below) can define additional
symbols to go into the nocache section
. = ALIGN(4);
KEEP(*(NonCacheable))
Signed-off-by: Mahesh Mahadevan <mahesh.mahadevan@nxp.com>
Fixes: #48950
Kconfig requires quoted strings in its configuration files, like this:
> CONFIG_A_STRING="foo bar"
But CMake requires expects that strings are without additional qoutes,
and therefore qoutes are stripped when loading Kconfig config filers
into CMake.
This is particular important when the string in Kconfig is a path to a
file. In this case, not stripping the quotes leads to an error as the
file cannot be found.
When users pass a string to Kconfig through CMake, they are expected to
pass it so that qoutes are correct seen from Kconfig, that is:
> cmake -DCONFIG_A_STRING=\"foo bar\"
In CMake, those qoutes are written as-is to Kconfig extra config file,
and then removed in the CMake cache.
After Kconfig processing, the Kconfig settings are read back to CMake
but without quotes. Settings that was passed through the CMake cache,
for example using `-D` are written back to the cache, but this time
without the qoutes. This results in Kconfig errors on sub-sequent CMake
runs.
Instead of writing the Kconfig value setting back to the CMake cache,
introduce an internal shadow symbol in the cache prefixed with `CLI_`.
This allows the CMake cache to keep the value correctly formatted for
Kconfig extra config creation, while at the same time keep existing
behavior for CONFIG_ symbols read from Kconfig.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Given that 3.6 is now several years old and the current and previous
Ubuntu LTS ship with 3.8 or later, upgrade the minimum required Python
version from 3.6 to 3.8.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
This commit introduces a dedicated unit testing board.
Today, a dedicated Zephyr unit testing scheme exists but is different
from how a Zephyr build generally works.
For example Kconfig is not possible, resulting on various different
hacks to pass Kconfig settings from test cases / testcase.yaml through
CMake to the code.
Some directly as compile definitions, some as header files with forced
inclusion on sources, some with wrapper flags which again results in
different define being enabled. There is even cases where a second
forced header inclusion undefines previous defines.
Unit test often does a manual check for the right boards, like this:
> if (NOT BOARD STREQUAL unit_testing)
> message(FATAL_ERROR "This project can only be used with...")
> endif()
Introducing a dedicated unit_testing board under `tests/root` allows
us to use Kconfig in unit test samples, and thus proper `prj.conf` and
extra Kconfig fragments.
Generation of autoconf.h so the overall architecture follows regular
Zephyr builds.
Proper and uniform error messages when invalid board is selected.
The unit_testing board and arch is located under: `subsys/testsuite` so
that it is only available when find_package(Zephyr COMPONENTS unittest)
is used, and not available for regular Zephyr builds.
Kconfig generates autoconf.h which is applied as compile flag to
test binary which means that kconfig defines placed in ztest.h can now
be removed.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
The Zephyr CMake module kconfig.cmake will create files inside
KCONFIG_BINARY_DIR.
In a default Zephyr build setup, this directory is created by Zephyr
CMake module zephyr_module.cmake. But Zephyr CMake modules are modular
and thus it must be possible to use kconfig module without
zephyr_module.
Thus kconfig.cmake must set and create KCONFIG_BINARY_DIR if it does not
already exists.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Remove ZEPHYR_BASE as default root in list_boards.py.
This allows list_boards.py to be used to only print boards at given
root(s) without printing Zephyr default boards.
Secondly it remove the need in list_boards.py to have any knowledge of
ZEPHYR_BASE.
The `west boards` command already has ZEPHYR_BASE knowledge and can
easily add ZEPHYR_BASE to the list of roots it is already constructing,
thus removing the need for knowing that ZEPHYR_BASE is always added.
Update boards.cmake to pass ZEPHYR_BASE as an additional arch root, as
arch.cmake has not yet been processed,which means ZEPHYR_BASE is missing
in ARCH_ROOT list at this point in time.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
This reverts commit 87c6789355, restoring
commit 3b341085a2.
Restore the original change now that the underlying issue has been fixed
in 6cfb18686e.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Because the fixup files do not exist anymore, stop using "unfixed"
naming in favor of "generated".
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Devicetree fixup files existed previous to the current stable Devicetree
API. While they served their purpose, they are no longer necessary nor
used in-tree. This patch drops support for this legacy feature.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
This file is no longer used by device.h, so let's avoid spending time
generating the content formerly in it.
In order to leave a pointer in place for users who are expecting to
see it or are pulling it into their own builds, however, replace its
contents with an #error directive that tells them what happened. This
can be removed later on when we expect people are used to the
transition.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
ZEPHYR_SDK_INSTALL_DIR is set from the env if it is not passed
explicitly to cmake, but one usage was reading directly from the env,
not allowing passing the var to cmake instead.
Change find_package of Zephyr-sdk to read from cmake var not env var.
Signed-off-by: Jeremy Bettis <jbettis@google.com>
This commit allows a project to specify a given set of Kconfig targets
prior to sourcing kconfig.cmake.
This allows for greater flexibility when re-using kconfig.cmake in
other projects.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
This commit allows for specifying a dedicated Kconfig namespace, for
example: <namespace>_CONFIG
This namespace will then be used for handling loading of configuration
files into CMake configure scope, as well as handling of CMake cache
variables when set using `-D<namespace>_CONFIG_<var>=<value>`.
This allows greater flexibility when re-using the cmake/kconfig.cmake
in other projects.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Allow build system to define an alternative defconfig file.
This allows reusing the existing kconfig.cmake file in other projects,
such as multi image builds.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Introduce a generated Kconfig.dts that sets a Kconfig symbol for
every compatible.
* We set DT_HAS_<compat>_ENABLED if the devicetree has a node with
<compat> enabled. (status is okay...)
We can then use the Kconfig symbol in driver Kconfig to determine
if the driver should be available, and thus enabled by default.
Signed-off-by: Kumar Gala <galak@kernel.org>
Move scripts needed by the build system and not designed to be run
individually or standalone into the build subfolder.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Move scripts needed by the build system and not designed to be run
individually or standalone into the build subfolder.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Optional way for generating symbols file using binutils. This file will
be generated on demand for tests which will then be parsed by twister to
create an accurate test plan from the listed testcase/suites.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
DTS directories can come from multiple places. Some places like a user's
CMakeLists.txt can preserve symbolic links. Others like
`scripts/zephyr_module.py --settings-out` resolve symbolic
links.
list(REMOVE_DUPLICATES ...) does not know anything about directories;
only about strings. To fix it the de-duplication, resolve all symbolic
links in DTS_ROOT first using file(REAL_PATH ...).
Fixes this somewhat confusing error:
```
-- Found BOARD.dts: /home/proj/mymodule/boards/myboard/myboard.dts
devicetree error: both /home/proj/mymodule/dts/bindings/intel,stuff.yaml
and /disk1/home/proj/mymodule/dts/bindings/intel,stuff.yaml
have 'compatible: intel,stuff'
```
Also, print `gen_defines.py` arguments in case of an error so anyone can
see what failed and not just west+CMake experts.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
zephyr_library_compile_options() was attempting to create
unique hashes for compile options in order to prevent creating
them multiple times. However, it was only using the first
argument to create the hash, so if multiple libraries had
different compile options but the first line was the same,
the second set would be mistaken for the first set and would
actually be passed the first set during compilation instead
of its own set.
The fix should be to use the entire compile options argument
list to create the hash so they should only match if the entire
options list is exactly the same.
This is a continued fix for #43835
Signed-off-by: Nicholas Lowell <nlowell@lexmark.com>
Update the new API to use K_USER as the flags for both
CONFIG_USERSPACE and CONFIG_TEST_USERSPACE. Also, fix the linker
script to properly include the suites, tests, and rules.
Fixes#44108
Signed-off-by: Yuval Peress <peress@google.com>
The ztest unittest module relies on some headers that fake the real
ones, e.g. <zephyr/arch/cpu.h>. When new header location was introduced,
their path was not updates. This patch fixes this problem and provides
ztest with both, legacy and new include paths so that it can keep
working on both scenarios.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Update the new API to use K_USER as the flags for both
CONFIG_USERSPACE and CONFIG_TEST_USERSPACE. Also, fix the linker
script to properly include the suites, tests, and rules.
Fixes#44108
Signed-off-by: Yuval Peress <peress@google.com>
This PR allows the user to add symbols to the ramfunc
section. The use for this could be as follows:
zephyr_linker_sources_ifdef(CONFIG_ARCH_HAS_RAMFUNC_SUPPORT
RAMFUNC_SECTION
quick_access_code.ld
)
quick_access_code.ld (as shown below) can define additional
symbols to go into the ramfunc section
. = ALIGN(4);
KEEP(*(CodeQuickAccess))
Signed-off-by: Mahesh Mahadevan <mahesh.mahadevan@nxp.com>
Propagate the board revision to Kconfig via the environment.
This is useful for application code to have access to for similar
reasons that CONFIG_BOARD is useful.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Add a Kconfig to disable non prefixed includes. By setting
`CONFIG_LEGACY_INCLUDE_PATH=n` developers can disable having
`include/zephyr` in the search path. This means that includes such
as `#include <kernel.h>` will no longer work.
Internally, every header should be updated to add the `zephyr/`
prefix to the headers. Only then, will developers be able to use
this config value for their applications.
Signed-off-by: Yuval Peress <peress@google.com>
Move include paths and add new target_include_directories to support
backwards compatibility:
* /include -> /include/zephyr
example: <irq.h> -> <zephyr/irq.h>
Issue #41543
Signed-off-by: Yuval Peress <peress@google.com>
This patch is doing two things:
- it is removing the fallback on the path. This is not possible anymore
since the DT binding file is now actually requiring the
'zephyr,memory-region' property to be present from which the region
name is obtained.
- it is sanitizing the name when CONFIG_CMAKE_LINKER_GENERATOR is used
or not.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
This commit updates the Zephyr build system to use the Zephyr SDK by
default for all host operating systems, when `ZEPHYR_TOOLCHAIN_VARIANT`
environment variable is not explicitly set.
Note that the Zephyr SDK is now available on all three major host
operating systems (i.e. Linux, macOS, Windows), and there is no reason
to exclude this behaviour for non-Linux hosts anymore.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
COPY/NOCOPY flag was added to the end of the string with ':'
as separators. The python script then split the line on ':'
which breaks on Windows build because the string was
[MEM]:[FILE]:[COPY]
In windows you will have 'c:\' in the file path so the
line.split() in script will split on the 'c:'.
Move the COPY/NOCOPY to:
[MEM]:[COPY/NOCOPY]:[FILE]
Note that the comments at top of gen_relocate_app.py also
indicates this format.
Fixes#43950
Signed-off-by: David Leach <david.leach@nxp.com>
Currently when a node has a 'zephyr,memory-region' compatible and a
'zephyr,memory-region' string property, a new memory region is created
in the linker script.
Having a memory region without a section to place variables in could be
not that useful. With this patch we extend the memory-region mechanism
to also create sections.
The user can then place variables in the sections as usual by using for
example the GCC attributes.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
Fixes: #43835
In zephyr_library_compile_options() the existence of the compile
options interface library is checked and function returns if it already
exists. This results in #43835 meaning two libraries cannot add the
same option.
This commit fixes this by re-using the already created unique interface
library and link this to the Zephyr library.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Although CMake scripts already support duplicated module
names, if two modules with the same name provide different
Kconfig files, then both files will be loaded leading to
potential conflicts.
Modify zephyr_module.py to enforce that all modules are
uniquely named so that it is possbile to override some of
the built-in modules using ZEPHYR_EXTRA_MODULES variable.
Signed-off-by: Damian Krolik <damian.krolik@nordicsemi.no>
Fixes: #43378
Move the exporting of ZEPHYR_TOOLCHAIN_VARIANT from the custom target to
the COMMON_KCONFIG_ENV_SETTINGS variable.
This ensures that the setting is exported both when running the initial
kconfiglib parsing but also on later menuconfig / guiconfig invocations.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
The STRUCT_SECTION_ITERABLE have been renamed to drop the Z_ prefix,
this drops the last reference in the cmake comments.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Don't translate from relative to absolute include paths.
It is lost to history why this was necessary and it is causing
problems with generator expressions so we remove the logic.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
Fixes: #43099
The CMake 3.22.1 / 3.22.2 PyPI version suffers a bug in the
`cmake_path(... PARENT_PATH)` implementation.
Therefore, when CMake version 3.22.1 / 3.22.2 is detected, test if the
CMake version is suffering from the bug, and in case the bug is present,
fail with an error regarding the issue.
The reason for failing, and not implementing work arounds is that Zephyr
already uses `cmake_path()` at two locations, and we cannot prevent
contributors from adding code which uses this function.
Secondly, Zephyr modules may also use `cmake_path()`, and thus be
affected by said bug.
It is impractical to implement work arounds at all possible locations
for something that is a CMake bug.
Therefore the safest solution is to test CMake itself, to check if the
version in use suffers said bug, and fail with a proper error message
if an affected CMake version is used.
See more here:
https://gitlab.kitware.com/cmake/cmake/-/issues/23187https://github.com/scikit-build/cmake-python-distributions/issues/221
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Instead of setting XCC_USE_CLANG=1, this patch adds xcc-clang toolchain
that is basically xcc using the clang compiler.
Initially, the new toolchain simply includes files from current xcc
toolchain and (re)sets some variables. This should be a more scalable
approach to diverge the toolchains in the future than placing
"if($ENV{XCC_USE_CLANG})" at several places.
It should also help to filter tests that run (or not) exclusively with
the clang variant of XCC on twister.
The XCC_USE_CLANG flag is documented as deprecated, and a message is
emitted during build if still in use. Its new behaviour is to instruct
Zephyr to use `xcc-clang` toolchain.
Signed-off-by: Ederson de Souza <ederson.desouza@intel.com>
This commit allows to load only subcomponents of the default Zephyr
CMake module, zephyr_default.cmake.
This allows other tools to execute Zephyr CMake build system up to a
specific module and the stop further processing.
This commit is an enabler for future support in twister to process only
devicetree or kconfig, to allow test / sample filtering before
generating a complete build system.
Sub-components can be loaded as:
> find_package(Zephyr COMPONENTS zephyr_default:<sub-component>)
for example:
> find_package(Zephyr COMPONENTS zephyr_default:dts)
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Creating a doc.cmake to the new Zephyr CMake modules dir.
This removes the need for `set(NO_BOILERPLATE TRUE)` before loading the
Zephyr CMake package.
It also removes the need within the doc/CMakeLists.txt file to manually
include individual parts of the Zephyr CMake files as this is now
controlled through a single Zephyr CMake doc module.
This aligns the way a Zephyr package is sourced with other places.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Move the unittest.cmake to the new Zephyr CMake modules dir.
This allows us to have a single Zephyr CMake package and load unittest
module as: 'find_package(Zephyr COMPONENTS unittest)'
This unifies the way Zephyr package is sourced and removes the need for
a dedicated ZephyrUnittest package.
Deprecate the use of: 'find_package(ZephyrUnittest)'
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Create a cmake/modules folder containing all Zephyr CMake modules.
All Zephyr cmake files that are included from boilerplate are now
converted into CMake modules which can be individually loaded.
The Zephyr CMake package is updated to support loading of individual
CMake modules using the COMPONENTS argument to `find_package(Zephyr)`.
If the COMPONENTS argument is not specified, the default Zephyr build
system will load.
If COMPONENTS is specified then, only those components and the
dependencies will be loaded.
If a Zephyr CMake module depends on another CMake module which has not
been loaded, it will automatically be loaded.
This allows us to modularize and reuse individual parts of the Zephyr
CMake build system in a more flexible way in future.
Such usage could be:
- Higher livel multi image build system
- Invocation of individual components, for example dts processing by
twister without loading all build code
- Doc build
- Unittesting
With this new CMake package and CMake module scheme then direct
sourcing of boilerplate.cmake has been deprecated.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
