I've repeatedly seen that people are not aware of the existence of
this configuration option.
I've been using build.pristine=auto daily for years and never had any
problems. I've also asked around on Slack a couple of times over
various points to see if anybody objects to making this change. Nobody
has, so let's just turn it on by default.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This commit message is a bit of a novel mostly:
- because the issues involved are longstanding
- as evidence this is not a capricious refactoring
The runners.core.RunnerConfig Python class holds common configuration
values used by multiple runners, such as the location of the build
outputs and board directory.
The runners code, first written in 2017-ish, replaced various shell
scripts that got this information from the environment. Avoiding
environment variables was a requirement, however. It's ghastly to set
environment variables for a single command invocation on Windows, and
the whole thing was part of a larger push to make Zephyr development
on Windows better.
I had a hammer (the argparse module). Finding a replacement naturally
looked like a nail, so the information that ends up in RunnerConfig
got shunted from the build system to Python in the form of 'west
flash' / 'west debug' command line options like '--board-dir',
'--elf-file', etc.
I initially stored the options and their values in the CMake cache.
This was chosen in hopes the build system maintainer would like
the strategy (which worked).
I knew the command line arguments approach was a bit hacky (this
wasn't a nail), but I also honestly didn't have a better idea at the
time.
It did indeed cause issues:
- users don't know that just because they specify --bin-file on the
command line doesn't mean that their runner respects the option, and
have gotten confused trying to flash alternate files, usually for
chain-loading by MCUboot (for example, see #15961)
- common options weren't possible to pass via board.cmake files
(#22563, fixed partly via introduction of runners.yaml and the west
flash/debug commands no longer relying on the cache)
- it is confusing that "west flash --help" prints information about
openocd related options even when the user's board has no openocd
support. The same could be said about gdb in potential future use
cases where debugging occurs via some other tool.
Over time, they've caused enough users enough problems that
improvements are a priority.
To work towards this, put these values into runners.yaml using a new
'config: ...' key/value instead of command line options.
For example, instead of this in the generated runners.yaml file:
args:
common:
- --hex-file=.../zephyr.hex
we now have:
config:
hex_file: zephyr.hex
and similarly for other values.
In Python, we still support the command line options, but they are not
generated by the build system for any in-tree boards. Further work is
needed to deprecate the confusing ones (like --hex-file) and move the
runner-specific host tool related options (like --openocd) to the
runners that need them.
Individual board.cmake files should now influence these values by
overriding the relevant target properties of the
runners_yaml_props_target.
For example, instead of:
board_runner_args(foo "--hex-file=bar.hex")
Do this:
set_target_properties(runners_yaml_props_target PROPERTIES
hex_file bar.hex)
This change additionally allows us to stitch cmake/mcuboot.cmake and
the runners together easily by having mcuboot.cmake override the
properties that set the hex or bin file to flash. (The command line
arguments are still supported as-is.)
Combined with 98e0c95d91ae16f14e4997fb64ccdf0956595712 ("build:
auto-generate signed mcuboot binaries"), this will allow users to
build and flash images to be chain loaded by mcuboot in a way that
avoids calling 'west sign' and passing 'west flash' its output files
entirely.
While we are here, rename runner_yml_write to runners_yaml_append().
This function doesn't actually write anything, and we're here
refactoring this file anyway, so we might as well improve the
situation while we're at it.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
The 'runner_config' variable name is particularly misleading because
there is a class called RunnerConfig, and that variable does not
contain one.
Rename it to 'runners_yaml' since it contains the parsed contents of
the runners.yaml file. Rename the variable that refers to the path
itself to 'runners_yaml_path'. No functional changes expected.
This is prep work for redoing how actual RunnerConfig objects get
made.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Before venturing off into refactoring land, let's make sure we have a
type checker running on the main APIs used by the various runners so
we don't miss anything obvious.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Use mypy to type check the runners package.
The test procedure is now annoying enough to replicate locally that
I'm going to wrap it in a script. Do this for both UNIX and Windows
environments by writing that script in Python.
Keep the GitHub workflow up to date so we now get mypy results in CI.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This is needed for sanitycheck hardware maps which take the serial
number directly from USB metadata.
Signed-off-by: Øyvind Rønningstad <oyvind.ronningstad@nordicsemi.no>
When MCUBoot is built using the swap-move strategy, the secondary slot
can be one sector smaller than the primary slot, because the primary
slot's last "useful" image sector must be "reserved" for the move
operation; this impacts the generation of padded images, so when
signing an image, the proper size to use is the secondary slot's,
unless a secondary image is not defined which is the case when using
single image DFU.
Signed-off-by: Fabio Utzig <fabio.utzig@nordicsemi.no>
Add support for the J-Link Commander "-NoGui 1" command line parameter
in the West J-Link runner.
This command line parameter suppresses GUI dialogs (except for license
dialogs) in J-Link Commander starting from v6.80.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
If a runner supports the --erase option, make it so that the user can
add --no-erase to the command line to explicitly disable it.
Add a diagnostic at info() level whenever this option is requested.
The intent of this commit is to make it more obvious when a mass erase
was requested, especially in situations (like MCUboot with
CONFIG_MCUBOOT_TRY_MASS_ERASE=y) where this option may be turned on by
default.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Fixes: #27375
This is a cleanup of the Zephyr CMake package export.
The code has been simplified so that the export now happens through a
CMake script. This avoids several generated CMake build files compared
to previous export mode, and thus removes the need for a CMake pristine
script.
A benefit of this cleanup is that it also fixes#27375.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Avoid use of the cached_runner_config() helper, which we have a TODO
item to get rid of. Adjust the output and do some other minor
improvements.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Rolling yet another parser turns out to be the best way to let west
extensions respond to Kconfig values. See source code comments in the
patch for details.
The sanitylib library has some similar functionality but it isn't
exactly the same, and it gets strings wrong. For example, that parser
can't handle this option:
CONFIG_FOO="he said \"no\" to me"
This one can, and it has a couple of other features we'll find useful
for west extensions eventually besides.
(Not to mention that sanitylib also rolled its own CMake cache parser,
which also exists in west_commands.)
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Consolidate creation of edtlib.EDT objects from a build directory's
devicetree into one place by loading it from build/zephyr/edt.pickle
everywhere. A previous commit creates edt.pickle from gen_defines.py.
In addition to probably speeding things up slightly by not reparsing
the devicetree, the main benefit of this approach is creating a single
point of truth for the bindings directories and warnings
configuration, meaning we don't have to worry about them getting out
of sync while being passed around between devicetree creation and
usage time.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
The `TEXT_SECTION_OFFSET` symbol is used to specify the offset between
the beginning of the ROM area and the address of the first ROM section.
This commit renames `TEXT_SECTION_OFFSET` to `ROM_START_OFFSET` because
the first ROM section is not always the `.text` section.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
When running CMake, then Python3 will be used.
This is detected through FindPython3, with a preference for using the
python or python3 in path, if any of those matches the required Python
minimal version in Zephyr.
It is also possible for users to specify a different Python, as example
by using:
`cmake -DPYTHON_PREFER=/usr/bin/python3.x`
However, when running `west` as native command, then west will be
invoked on linux based on the python defined in:
`west` launcher, which could be: `#!/usr/bin/python3.y`
Thus there could be mismatch in Pythons used for `west` and the python
used for other scripts.
This is even worse on windows, where a user might experience:
```
>.\opt\bin\Scripts\west.exe --version
Traceback (most recent call last):
File "C:\Python37\lib\runpy.py", line 193, in _run_module_as_main
"__main__", mod_spec)
...
File "C:\Python37\lib\socket.py", line 49, in <module>
import _socket
ImportError: Module use of python38.dll conflicts with this version of
Python.
```
when testing out a newer Python, but the python in path is still a 3.7.
By importing `west` into zephyr_module.py and by using, as example
`python -c "from west.util import west_topdir; print(topdir())"`
we ensure the same python is used in all python scripts.
Also it allows the user to control the python to use for west.
It also ensures that the west version being tested, is also the version
being used, where old code would test the version imported by python,
but using the west in path (which could be a different version)
If the west version installed in the current Python, and west invocation
is using a different Python interpreter, then an additional help text
is printed, to easier assist users with debugging.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
BOSSAC uses a bootloader, so pull the flash address from DeviceTree or
the config and use if the version of BOSSAC supports offsets.
Signed-off-by: Michael Hope <mlhx@google.com>
BOSSA 1.7 has built-in bootloader offset handling while 1.9.1
requires the user to supply the offset. Add support for both by
sniffing the help output and warn the user if the flags needs adding.
Related to #22062
Signed-off-by: Michael Hope <mlhx@google.com>
This commit creates a list of a phony targets for each runner, that is:
`west_flash_depends`, `west_debug_depends`, and so on.
Those targets has identical dependencies as CMake runner target.
flash, debug, debugserver, attach targets.
As example `ninja flash` correctly ensures dependencies are taken into
consideration before calling `west flash`.
Unfortunately, calling `west flash` directly would not re-run the flash
dependencies, cause `west flash` would only build the default CMake
target.
Now, `west flash` calls the phony `west_flash_depends` target, ensuring
all deps are up-to-date before flashing (unless --skip-rebuild is given)
The same is true for the other mentioned runners.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Just changes to the west help output; no functional changes expected.
Make option descriptions lowercase to match the argparse module's
conventions. When multiple sentences are required, move them to parser
prolog/epilog or argument group description sections.
Clarify some points that have confused multiple people.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Move the --erase option to core.py from nrfjprog.py and jlink.py,
where it is currently supported.
Using the RunnerCaps option enforcement mechanism introduced earlier,
enforce that it will only be given to runners that support it.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Require all implementations to provide a do_create(), a new
ZephyrBinaryRunner abstract class method, and make create() itself
concrete.
This allows us to enforce common conventions related to individual
runner capabilities as each runner provides to the core via
RunnerCaps.
For now, just enforce that:
- common options related to capabilities are always added, so runners
can't reuse them for different ends
- common options provided for runners which don't support them emit
sensible error messages that should be easy to diagnose and support
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
We can just call super() instead of super(MyClassName, self). The
original extra verbosity is likely due to old habits of mine from
Python 2 which are no longer necessary, but got copy/pasted around.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
The argparse module, by default, complains about non-defined
arguments. This prevents passing arguments prefixed with '-'
or '--' to the target script (e.g. calling another Python
script using argparse). This changes the misc-flasher script
so that any arguments not recognized by west will be passed
to the target script.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The start_addr attribute of Stm32flashBinaryRunner is an integer. It
must be converted to a string before being concatenated with a colon and
the (already converted to a string) size to erase or write.
Signed-off-by: Grzegorz Szymaszek <gszymaszek@short.pl>
Disables the Debug-Access-Port of the microcontroller after flashing.
If not disabled, the DAP consumes ~1.6mA until the debugger disables it
or a hard power cycle is applied.
The DAP is typically automatically disabled after flashing, but if other
instances of JLink software are running (not connected), it will not be.
The added command resets the value of the CTRL/STAT register of the DAP.
This clears the CSYSPWRUPREQ and CDBGPRWUPREQ bits, leaving the debug
hardware free to power off the appropriate hardware. In no way does it
hinder the ability to later connect to the device for debugging.
This resolves the jlink portion of #26139
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
* add the runner script for metaware debugger(mdb).
* mdb is required for SMP case
* mdb also can provides a GUI interface
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
Catch ImportError whenever a non-standard module import fails from any
runners that do one. Complain at runtime about it if the user actually
needs the runner.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
OSX compatibility added bossac runner.
Linux behvaiour is unchanged.
Window is now explicity mentioned as incompatible
Signed-off-by: Brendon Le Comte <brendon.lecomte@gmail.com>
In case of flashing (which is typically used with OpenOCD)
we do reset of the target after programming application binary
in the non-volatile memory.
In case of Elf execution we need to reset the target before
loading Elf sections so that we might be sure our target
is in sane & expected state before we start execution.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
The variables with the STATIC type aren't read.
The commit 877fc59e30 introduce the read of CMAKE_PROJECT_NAME
which could be defined in CMakeCache.txt file like that:
CMAKE_PROJECT_NAME:STATIC=MyProject
If STATIC type is not managed, the CMAKE_PROJECT_NAME isn't set and
CMake is always force to run again
Signed-off-by: Julien D'Ascenzio <julien.dascenzio@paratronic.fr>
Fix two issues:
1. The script assumes the default CMake generator build tool
platform is installed. On Linux at least, that's Make instead
of Ninja, but Make might not be installed since Zephyr recommends
Ninja. On Windows, that might be VS Code or nmake.
Calling `cmake -P pristine` instead of `cmake --build <path>
--target pristine` has the benefit of removing the dependency on a
build command, and hence the default generator is not relevant.
2. It also assumes run_cmake() returns control, and therefore pristine
can be run.
However, if the cmake command fails hard (say, due to issue #1
before this patch), run_cmake() throws an exception instead.
Fix that by trying to run the pristine target in a finally block
instead, and adding some manual cleanup steps in case the build
system is in a bad state and pristine fails too.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Signed-off-by: Torsten Rasmussen <torsten.rasmussen@nordicsemi.no>
If there is an error in the CMake configuration phase (this can happen
if a script run using execute_process() fails, for instance), the
build system is incompletely generated and future attempts to run
'west build' will fail. This manifests in the following error:
Error: could not find CMAKE_PROJECT_NAME in Cache
Whenever we see that the cache exists but this variable is missing,
let's just force CMake to run again. This avoids the error in my
testing and is a bit more user friendly. I've seen multiple users
asking what to do in this situation; the answer is always "just build
it again", so we might as well do it for them.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
The board name for the nRF52 DK, so far known as nrf52_pca10040, is
renamed to nrf52dk_nrf52832. Its documentation and all references
to its name in the tree are updated accordingly. Overlay and
configuration files specific to this board are also renamed, to
match the new board name.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Fixes: #23872
Relocating Zephyr Unittest CMake package to ensure that
HINTS ${ZEPHYR_BASE} in
find_package(ZephyrUnittest HINTS ${ZEPHYR_BASE}) works correctly when
the package has not been exported to CMake user package registry.
This ensure that the new package functionality is fully backwards
compatible on systems where the package is not exported and ZEPHYR_BASE
is set.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
This commit includes the following fixes in order to remove environment
setting of ZEPHYR_BASE is west extension commands.
- Build command
west build --pristine will now use the ZEPHYR_BASE variable found
in CMakeCache.txt in the build folder.
This ensures that the pristine command is executed from the same
Zephyr that was used for compilation.
- Board command
The west boards command no longer sets Zephyr base before invoking
cmake -P cmake/boards.cmake
Instead boards.cmake uses find_package(Zephyr) to ensure consistent
behavior with Zephyr samples, so that the detection of Zephyr base is
uniform across CMake commands.
It also changes BOARD_ROOT_SPACE_SEPARATED to BOARD_ROOT in order to
be consistent with existing user documentation.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This command make it possible to use west for first time registration
of a Zephyr config package in CMake.
To register Zephyr as a CMake config package, invoke:
west zephyr-export
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Originally reported in #23539 (though that seems to have been another
problem), west flash and friends are dumping stack when used with an
unconfigured runner.
Let's just promote the warning about this to an error. The idea that
this ever could have worked without explicit support has not worked
out in practice, to my knowledge.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
The header printed at the beginning of every west build is kind of
annoying. Let's remove it and just print the application source
directory at cmake time instead. The build directory and board are
already printed there, anyway, and that's all IDE users will see.
Let's clean up the BOARD to make it say "board" instead. That matches
the west build --board option name a bit more closely and is still
legible.
Likewise, let's not print any west build messages if we're just
incrementally recompiling. That's noisy.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Martí Bolívar
Øyvind Rønningstad
Fabio Utzig
Henrik Brix Andersen
Torsten Rasmussen
Andrei Emeltchenko
Maureen Helm
Stephanos Ioannidis
Michael Hope
Daniel Leung
Grzegorz Szymaszek
Jordan Yates
Wayne Ren
Brendon Le Comte
Alexey Brodkin
Julien D'Ascenzio
Carles Cufi