Make headers look like this instead:
/* From ${ZEPHYR_BASE}/arch/common/isr_tables.ld: */
Besides being shorter and more "private", this makes the content of
snippets-*.ld files the same no matter who built them and where.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
Allows snippets to be placed in different locations:
- The noinit, rwdata and rodata output sections
- Two different locations for placing custom output sections,
one location for RAM and another for all other sections.
Signed-off-by: Øyvind Rønningstad <oyvind.ronningstad@nordicsemi.no>
When importing a pre compiled imported library it is currently
required to perform three steps.
This commit introduces a helper function which allows the
user to import a library with a single function call.
Signed-off-by: Håkon Øye Amundsen <haakon.amundsen@nordicsemi.no>
The function zephyr_append_cmake_library works with
libraries that does not have source files.
This change removes an incorrect statement related
to that.
Signed-off-by: Håkon Øye Amundsen <haakon.amundsen@nordicsemi.no>
While this race seems unlikely and harmless let's play it safe and
implement the usual solution: temporary file + atomic rename.
The race is unlikely and maybe even harmless because:
- Only sanitycheck seems to invoke cmake concurrently.
- Users rarely delete their ~/.cache/zephyr/ToolchainCapabilityDatabase/
- All concurrent cmake processes write the same, single byte to the same
files.
- Creating a single byte is at least very fast, so extremely short
window for others to read an empty file.
For additional background see links in issue #9992
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
Update the files which contain no license information with the
'Apache-2.0' SPDX license identifier. Many source files in the tree are
missing licensing information, which makes it harder for compliance
tools to determine the correct license.
By default all files without license information are under the default
license of Zephyr, which is Apache version 2.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
PR #14776 / commit 915a353255 changed function
zephyr_library_compile_options() to use MD5 instead of RANDOM for build
reproduction reasons. As later predicted by Sebastian Bøe in the PR
(thx), the names generated for these pseudo-libraries won't be unique
anymore if the exact same flag gets added more than once.
To reproduce the issue, simply add the following two lines in some
CMakeLists.txt file like samples/hello_world/CMakeLists.txt:
zephyr_library_compile_options("-Dfubar=barfu")
zephyr_library_compile_options("-Dfubar=barfu")
cmake will then fail like this:
CMake Error at zephyr/cmake/extensions.cmake:403 (add_library):
add_library cannot create target
"options_interface_lib_e689fdb7eb15d801c2f95dd61301fc5e" because
another target with the same name already exists. The existing
target is an interface library created in source directory
"zephyr/samples/hello_world". See documentation for
policy CMP0002 for more details.
Call Stack (most recent call first):
CMakeLists.txt:9 (zephyr_library_compile_options)
As requested by Sebastian, silently discard duplicate options just like
CMake does.
Fixes#15093
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
1. To support being called multiple times, the function
zephyr_library_compile_options() uses unique
options_interface_lib_${random} names. These only need to be unique, not
random. So replace them with a simple MD5 hash of the ${item} argument.
To reproduce quickly run:
sanitycheck -b -p qemu_xtensa --tag shell
... a few times and compare the output directories.
This bug sits for now at the top of the list of build reproducibility
issues the most bizarre and difficult to root cause. When running
sanitycheck over a large sample of configurations it was affecting
only qemu_xtensa/**/zephyr/arch/arch/xtensa/core/startup/Makefile
files (and their corresponding CMakeFiles/TargetDirectories.txt),
randomly ordering the following three Make targets and only these
three: rebuild_cache, edit_cache, arch__xtensa__core__startup.
The key to root causing was cmake --trace-expand which prints the random
string.
2. generate_unique_target_name_from_filename() also generated a random
string for the same uniqueness reason. This one was easier to root cause
and reproduce with: sanitycheck --tag gen_inc_file
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
We'd get output from a cmake build that would have lines like:
Version: DTC 1.4.7
Version: DTC 1.4.7
This was a side effect of check_dtc_flag. Add the OUTPUT_QUIET option
to execute_process to supress this output.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Add a function that let's us check if a given dtc flag can be used with
the version of dtc we have. This lets us utilize specific warning flags
that might not exist in older dtc versions and still have things work
properly.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Rather than specifying SHIELD via Kconfig, we move it to being
specified via the command line, similar to board.
So we can do:
-DSHIELD=x_nucleo_iks01a1
or, for multiple shields:
-DSHIELD="x_nucleo_iks01a1 frdm_kw41z"
Following cmake change, update x_nucleo_iks01a1 sample in order
not to enable CONFIG option anymore but set SHIELD cmake option.
Last, update documentation to reflect this change.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Cosmetics, no functional change expected.
Fixed leading space alignment.
Replaced tabs with spaces.
Emulation error message output is now aligned.
To locate tabs in cmake, the following bash is useful:
grep -PRil "\t" * | grep -i cmake | grep -v ^sanity
Signed-off-by: Mark Ruvald Pedersen <mped@oticon.com>
CheckCCompilerFlag and CheckCXXCompilerFlag does not belong to
extensions.cmake since they are standard CMake files. Boilerplate
seems more appropriate.
Signed-off-by: Mark Ruvald Pedersen <mped@oticon.com>
Generalize 'import_kconfig' to support parsing symbols with other
prefixes than 'CONFIG_'. This will eventually allow us to also import
'DT_' symbols into the CMake namespace.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
This patch creates a rule in the cmake to trigger the generation
of linker_relocate.ld and code_relocation.c files.
The linker_relocate.ld will create appropriate sections and will
link the required functions or variables from all the selected
files.
The code_relocation.c will have code that is needed for
initializing data sections and copy of text sections(if XIP).
Also this will contain code that is needed for zeroing of bss.
The procedure to invoke this feature is:
1. Enable CONFIG_CODE_RELOCATION in the prj.conf
2. Inside CMakeList.txt in the project we need to mention
all the files that needs to get relocated.
zephyr_kernel_code_relocate(src/*.c SRAM2)
Where the first argument is the file/files and the second
argument is the memory where it has be placed.
NOTE: The file argument supports glob expressions.
NOTE: Step 2 can be done as many times as required. And relative
paths can be given here.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
CMake prints a helpful error message with all the boards supported
when a board name has been misspelled. But CMake also
remembers (caches) the corrupted board name and refuses to change it
until the board directory has been cleared.
CMake is rightly hesitant to change the board, as other cached
variables depend on what the board is. But it is safe to change the
board just after it has been selected, because we verify it's
correctness before using it to calculate other variables.
So to support the use-case of changing a misspelled board
name (without clearing the build directory) we unset the cached board
when it is invalid.
Also; change the now unused (in-tree) macro 'assert_with_usage' to
'print_usage'.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
The CMake extension 'generate_inc_file_for_target()' has some code for
generating a unique target name that is needed elsewhere, for instance
by DeviceTree.
Make the code re-usable by refactoring it out into it's own function.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
CI sometimes fails with a temporarily corrupted toolchain capabilitiy
database file. Although not proven, there is evidence that CMake's
file(APPEND does not work atomically when there are concurrent writes
and reads of a certain size.
To avoid file appending, we re-write the key-value database
implementation to store keys in filenames and values in individual
single-byte files.
This is (most likely) fixes#9992.
NB: Users that have been overriding the database file location with
the CMake variable ZEPHYR_TOOLCHAIN_CAPABILITY_CACHE must now specify
a directory with the variable ZEPHYR_TOOLCHAIN_CAPABILITY_CACHE_DIR.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
It is very inconvenient to maintain an application that both runs on a
Zephyr board and an out-of-tree board.
It forces one to write build scripts like this in the app:
if(BOARD STREQUAL my_out_of_tree_board)
set(BOARD_ROOT some/out/of/tree/board/path)
endif()
To avoid this we change the semantics of BOARD_ROOT. Instead of it
being a path to the board root it is now a prioritized list of board
root directories. Zephyr will append ZEPHYR_BASE to BOARD_ROOT.
This ensures that Zephyr boards can be used when the out-of-tree board
directory can not supply the requested board.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
zephyr_library_compile_options will apply compile options to a library
with high priority. Due to bug #8438 these compile options have been
applied not only to the library, but also to it's public interface and
therefore to any library that has been linking with the library.
To resolve#8438 we use PRIVATE linking instead of PUBLIC linking.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
CMake has several prototypes/signatures for the function
'target_link_libraries'. This commit migrates the usage of
'target_link_libraries' on Zephyr CMake libraries from the old 'plain'
signature to the new '<PRIVATE|PUBLIC|INTERFACE>' signature.
For technical reasons the two signatures can not be mixed. Each
library must exclusively use either the old or new signature.
The 'old' plain signature is equivalent to using the PUBLIC
signature. Migrating to use 'PUBLIC' is therefore expected to be a
safe change.
After the migration it will be possible to use the PRIVATE and
INTERFACE signatures on Zephyr CMake libraries. This is useful for
instance to fix issue 8438.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
This cmake macro was changing the control flow of the cmake files.
It is not advisable to modify the callee control flow from inside
a cmake macro.
Hence removing all instances of this macro from the build system.
Fixes: GH-7565
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
This is a stepping-stone to adding runner functionality into west
itself.
Since all of the runner tools assume a Zephyr build directory layout,
this doesn't put anything generic into a Zephyr-specific tool.
Make minimal adjustments to zephyr_flash_debug.py to keep existing
build system targets working unmodified.
Signed-off-by: Marti Bolivar <marti@opensourcefoundries.com>
Persist all the important information needed by the runner package to
the CMake cache. This serves as a parseable record for various tools
which need to understand how to run the binary.
In particular, it will be used by the west tool, which will be
introduced in subsequent patches.
Signed-off-by: Marti Bolivar <marti@opensourcefoundries.com>
Add a variant on generate_inc_file_for_target in which the user can
specify the target that the generated file will depend on. This allows
exposing that target to other possible places where a dependency on the
generated file existing might be needed.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Checks if a certain Kconfig is enabled and then do the same
operation as zephyr_library.
If it finds that the Kconfig is not set it will not
create a new library and will stop processing the rest of
the CMakeFile.txt. Usefull in processing the driver Cmake
files.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
When a bug in the caching mechanism is detected the cache format must
be bumped to ensure that we no longer get a cache hit on old corrupted
keys.
This fixes#7246 when the cache is corrupted.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
Messages about compiler checks are very verbose so we disable them by
default. They have always been undesirably verbose, but introducing
CMAKE_REQUIRED_FLAGS into the 'check' string made them unacceptably
so.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
Fix a bug where different compiler checks were aliased and therefore
the test results were incorrectly re-used.
The 'check' string is used by CMake internally to cache test results,
but when testing linker flags the check string has been aliased
between the different linker checks. To fix the aliasing issue we add
the CMAKE_REQUIRED_FLAGS variable to the 'check' string. The aliasing
issue disappears because the linker flag-under-test is in
'CMAKE_REQUIRED_FLAGS'.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
This commit ports nearly all usage of check_c_compiler_flag,
check_cxx_compiler_flag, and check_compiler_flag to use
zephyr_check_compiler_flag instead.
This has a significant CMake configure-time performance impact.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
zephyr_check_compiler_flag is introduced to resolve
https://github.com/zephyrproject-rtos/zephyr/issues/5633
The time spent when invoking CMake is halved with a warm cache.
See the API documentation for details.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
This commits allows check_compiler_flag to accept the empty string as
an option. There are several use-cases for accepting no flag, one of
which is testing the toolchain itself.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
To improve the build system's "clean build" runtime we need to cache
files in a directory outside of the build directory.
Caching files outside of the build directory is is already being done
to a certain degree today. For instance, if a user has installed
ccache it will cache files in $HOME/.ccache/.
This commit locates a directory that is suitable for caching files in
a cross-platform way. For instance on Linux this is believed to be
$HOME, on Windows this is believed to be %LOCALAPPDATA%.
If for whatever reason no environment variables are are found to be
suitable we fall back to using $ZEPHYR_BASE/.cache/. For users that
often use 'git -clean' the caching mechanism will not work as well,
but it will at least be better than no caching at all.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
Added a CMake extention for determining if a directory is write-able
by the build system. To determine this, we attempt to 'touch' a file
in the directory and check the return code of this command.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
Fixes issue #7184.
Use $BOARD_ROOT to find correct list of supported boards.
This is needed since the user can set this variable to any directory.
Pass $BOARD_ROOT to usage.cmake where it is called.
Signed-off-by: Håkon Øye Amundsen <haakon.amundsen@nordicsemi.no>
We are using the CMake command 'file(STRINGS' (which defaults to only
decoding ASCII) to read Kconfig configurations. When UTF-8 characters
are detected they are treated as newlines.
This behaviour has caused issues when users have input UTF-8
characters into Kconfig values. E.g. USB device descriptor strings.
This commit adds the flag 'ENCODING "UTF-8"' to the 'file(STRINGS'
command so that UTF-8 characters are correctly passed through the
build system.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
We want to support other toolchain not based on GCC, so the variable is
confusing, use ZEPHYR_TOOLCHAIN_VARIANT instead.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Error occurred when using target_link_libraries in CMakeLists.txt:
Cannot specify link libraries for target "interface" which is not built
by this project.
Replace "interface" with "${interface}".
Signed-off-by: Ding Tao <miyatsu@qq.com>
Due to a chicken-and-egg problem, it's not possible to call
board_runner_args() directly from an application CMakeLists.txt and
have it take effect:
- if the board_runner_args() call appears before including the
boilerplate file, then the function is undefined and an error
occurs.
- if the call appears after including the boilerplate file, then
board_finalize_runner_args() has already been called by the
relevant runner.board.cmake file, so it won't take effect.
Similarly to the way set_conf_file() works, allow applications to
declare additional runner args by checking if they have provided a
macro named app_set_runner_args(), which contains calls to
board_runner_args(). Using a macro means that we can invoke it with
the definition of board_runner_args() in scope.
Signed-off-by: Marti Bolivar <marti@opensourcefoundries.com>
Instead of accessing the environment variable ZEPHYR_BASE every time we
require accessing the source code root, use an intermediate variable
that has OS path separators correctly set to '/' to avoid issues on
Windows.
Note: This removes the ZEPHYR_SOURCE_DIR CMake variable. External
applications using that will need to change to use the new ZEPHYR_BASE
variable.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Usually the zephyr_get_* API returns values prefixed with tokens like
-I -system, -D. But sometimes we need the values without these
prefixes, so we introduce the optional SKIP_PREFIX argument.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
Introduce Zephyr interface libraries to fix#5351.
From the documentation:
A Zephyr interface library is a thin wrapper over a CMake INTERFACE
library. The most important responsibility of this abstraction is to
ensure that when a user KConfig-enables a library then the header
files of this library will be accessible to the 'app' library.
This is done because when a user uses Kconfig to enable a library he
expects to be able to include it's header files and call it's
functions out-of-the box.
A Zephyr interface library should be used when there exists some
build information (include directories, defines, compiler flags,
etc.) that should be applied to a set of Zephyr libraries and 'app'
might be one of these libraries.
Zephyr libraries must explicitly call
zephyr_library_link_libraries(<interface_library>) to use this build
information. 'app' is treated as a special case for usability
reasons; a Kconfig option (CONFIG_APP_LINK_WITH_<interface_library>)
should exist for each interface_library and will determine if 'app'
links with the interface_library.
This API has a constructor like the zephyr_library API has, but it
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
The generate_inc_file_for_target() extension is a useful wrapper
around the more low-level generate_inc_file(). It takes care of
ensuring that the generated file correctly depends on it's source.
To do so it needs to name and define a custom_target, the name must be
unique, so it was constructed based on the path to the generated file,
but this caused multiple issues.
The intention was that the path of the file would help identify the
target, in hindsight, identifying the target has proven to not be
important.
This patch fixes#5466 by using a random string to construct the
target name instead of the path to the generated file.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
target_ld_options() was only testing whether a flag could be passed to
the compiler driver when it was compiling an object file. Not whether
a flag could be passed to the compiler driver when it was linking an
elf file.
For most flags, these tests would have the same result, but it does
not for flags like -Wl,--print-memory-usage or -Wl,notaflag.
This patch fixes#5488 by re-using the method from #5459.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
To avoid warnings from cmake such as:
'path' is reserved or not valid for certain CMake features, such as
generator expressions, and may result in undefined behavior.
This can be seen in Jenkins for example where @ is used when multiple
jobs are run on the same machine.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
The various runners (flash/debug scripts) use environment variables to
take arguments. This is legacy behavior which is not desirable.
Use command line arguments instead.
Note: this leaves more general environment variables with publicly
documented behavior in place for now, for compatibility, e.g.:
ZEPHYR_FLASH_OVER_DFU, OPENSDA_FW, ESP_IDF_PATH, PYOCD_DAPARG
For example, when using dfu-util to flash arduino_101, instead of
setting DFUUTIL_PID, DFUUTIL_ALT, and DFUUTIL_IMG environment
variables, have the script invocation look like this:
python3 .../zephyr_flash_debug.py dfu-util flash \
[common arguments omitted] \
--pid=8087:0aba --alt=x86_app \
--img=.../build/zephyr/zephyr.bin
Make similar changes for other runners (openocd, etc.) and
targets (debug, debugserver).
To implement this in the scripts:
- have the individual scripts/support/runner/some-runner.py files
register their own command line arguments
- teach them to construct instances from arguments, not the
environment
- have zephyr_flash_debug.py request runners to register command
line argument parsers, and handle arguments
In the build system:
- add a new board_runner_args() extension function that board.cmake
files can use to add to the zephyr_flash_debug.py command line
- adjust cmake/flash/CMakeLists.txt to invoke with arguments
- add new helper include files for each runner (like
boards/common/dfu-util.board.cmake, etc.), which add default
options as needed and then add on overrides from
board_runner_args() calls
- update board.cmake files to use the new includes and extension
This implied some tweaking when using openocd to make the CMake string
escaping and unescaping work properly.
Signed-off-by: Marti Bolivar <marti@opensourcefoundries.com>
