Defer loading .config until we really need it, when we are flashing a
binary. Pre-emptively loading it is wasted effort if we're flashing a
.hex, which has been the default behavior when possible since
dcaabb860f ("west: runners: jlink: prefer .hex over .bin").
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Put it all in one log message rather than splitting it up.
This makes it look cleaner now that each log message is prefixed with
'runners.link:'.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
JLink versions like 'V7.0a' do not conform to PEP 440 version
conventions; the 'a' part is used by PEP 440 compliant versions for
alphas. It gets parsed to a legacy type by the packaging library,
which always is treated as a lower value when compared with a
conforming version string.
To fix, get the version from the shared library distributed with the
JLink tools. This has the side benefit of making the code work on
Windows. That's merely a nice to have for -nogui 1 detection for now,
but will be essential in the next commit.
Reported-by: Jake Mercer <jake.mercer@civica.co.uk>
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This will be used to deal with the Segger shared library in a portable
way in the jlink runner.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Make it easier to get a BuildConfiguration from runner code.
Stash the result so it only has to be computed once.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This option has existed since the beginning of the runners package,
which greatly predates the way DT is used in zephyr right now. It
never really worked the way I wanted it to but it's too much work to
fix it now. Try to improve the help a bit at least while I'm looking
at it again.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Copy a fix from test_nrfjprog.py to the other runner test suites. The
current code will enter an infinite recursion if you hit the path
where os.path.isfile is called, since it's been patched to
os_path_isfile_patch in the calling context. The fix is to cache the
'real' version in the parent scope and call it directly as a fallback.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Initialize `handle.device` to `None` so that the assert after the
matching has a chance to catch errors. Without this, a failed match will
raise an exception on the line above as it attempts to get a property
that doesn't exist.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
This adds dictionary based logging support. Dictionary based
logging is binary based where one big difference is that
static strings are stored as pointers instead of the whole
string. This results in reduced space requirements for
storing log messages in certain scenairos.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Breathe project has released version 4.29.1, a fix release for the
issues found in 4.29.0. Relax version requirements by just skipping the
buggy version and staying to compatible releases.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Due to the use of gperf to generate hash table for kobjects,
the addresses of these kobjects cannot change during the last
few phases of linking (especially between zephyr_prebuilt.elf
and zephyr.elf). Because of this, the gperf generated data
needs to be placed at the end of memory to avoid pushing symbols
around in memory. This prevents moving these generated blocks
to earlier sections, for example, pinned data section needed
for demand paging. So create placeholders for use in
intermediate linking to reserve space for these generated blocks.
Due to uncertainty on the size of these blocks, more space is
being reserved which could result in wasted space. Though, this
retains the use of hash table for faster lookup.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
There is no need to specify linker section for variables in
the output. The compiled object file from the produced
source file will be further processed to have them renamed
anyway.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
It's easier and safer to parse json file rather than extract
total size from stdout
Now the .json file has two keys:
{
"symbols": {
"children": [
...
],
"identifier": ":",
"name": "root",
"size": 2220
},
"total_size": 34272
}
Signed-off-by: Jingru Wang <jingru@synopsys.com>
Do calculation based on what is actually going to be run and evaluated
at runtime, ignore the cases we already know going to be skipped.
This fixes an issue where some sets would get majority of skips and
basically run nothing beside filtering.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
All skips for platforms listed in the .yaml's integration_platforms
will be treated as errors in the integration mode ('-G/--integration').
This feature serves to guarantee a proper testing scope in CI as no
integration platform will be skipped silently.
Fixes#33874
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
It can be convenient to "iterate" over the elements of a property, in
the same way it is convenient to "iterate" over enabled instances.
Add a new macro for doing this, along with a DT_INST_FOREACH_PROP_ELEM
variant.
This is likely to be more convenient than UTIL_LISTIFY or FOR_EACH in
some situations because:
- it handles inputs of any length
- compiler error messages will be shorter and more self-contained
- it is easier to use with phandle-array type properties, which
require more complicated macro boilerplate when used with
util_macro.h APIs
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
We already have support for handling the Zephyr binding "path" type in
edtlib.Node._prop_val(), but the binding inference code isn't making
use of that. Handle this type as well, as it is just as convenient as
Type.PHANDLE and can be more idiomatic depending on the situation.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Instead of hard-coding constants, use an IntEnum.
These is still a subclass of 'int', but is both easier to import and
easier to read during debugging.
For example, compare:
>>> Type.BYTES
<Type.BYTES: 1>
with:
>>> TYPE_BYTES
1
However, 'Type.BYTES == 1' is still True, and the enum values
otherwise behave like you would expect.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Add the ability to define architecture specific structures, notably
the ability to extend struct _cpu with per-CPU arch-specific stuff that
can be accessed with _current_cpu->arch.* similarly to _current->arch.*
for per-thead architecture data.
This is opt-in for architectures that want to benefit from this,
otherwise empty defaults are provided. A placeholder for ARM64 is
included to show the pattern.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Today, there is a build target is added for each runner: flash, debug,
debugserver, attach.
And those runners will have a dependency to Zephyr logical target that
is built before invoking `west <runner>`.
This design has some flaws, mainly that additional dependencies directly
on the target will not be built when running `west <runner>` directly.
That generator expressions cannot be used for the DEPENDS argument.
Instead, the build target `<runner>` will not have any dependencies, and
will raise a build error if a dependency is added to the target.
Due to how `add_dependencies()` work, this must be done as a build time
check, and not configure time check.
`west <runner>` will invoke a build before executing the runner, and
this way ensure the build target is up-to-date, which again removes the
need for a dedicated `west_<runner>_target`.
It also minimizes the risk of developer errors, as developers no longer
need to consider the need for adding additional dependencies.
If a custom target is part of the default `all` build, then it's ensured
to be up-to-date.
Fixes: Issue reported on slack.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
With the previous logic memory footprint can only be saved in a report
if a given test/sample was fully executed and passed (built and run),
hence build-only tests were not providing these metrics. This commit
modifies the logic so that it is enough to have the build successful
to be able to get the memory footprint.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
The info of rom/ram usage by an application was lost along the
way of data processing in twister. The commit add a line which
pass further these metrices as well.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
With the addition of #34185, it is not longer gauranteed that for a
memory region `NAME` there exists a symbol `NAME_ADDR`. Use the linker
builtin function `ORIGIN` to instead directly get the start address of
the selected memory region.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Removing CONFIG_TRACING_CPU_STATS in favor of
CONFIG_THREAD_RUNTIME_STATS which provides per thread stats. The same
functionality is also available when Thread analyzer is enabled with the
runtime stats enabled.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Upon discovery of a sample/testcase.yaml file, twister will disregard
further subfolders. Given the folder structure below, the testcase.yaml
will currently not be discovered:
samples/sample1/
samples/sample1/sample.yaml
samples/sample1/tests/
samples/sample1/tests/testcase.yaml
This is desirable to allow placing test code closer to sample code.
Signed-off-by: Jorgen Kvalvaag <jorgen.kvalvaag@nordicsemi.no>
The commit coccinelle/coccinelle@47bd4cae52 changed a behavior of
parsing right before coccinelle v1.1.0. With the commit, the current
scripts under scripts/coccinelle/ errors out with:
minus: parse error:
File ".../zephyr/scripts/coccinelle/deref_null.cocci",
line 25, column 42, charpos = 666
around = '...',
whole content = (E != NULL && ...) ? <+...E->f@p1...+> : ...
I've already raised an issue upstream coccinelle/coccinelle#257. But
Debian is already shipping v1.1.0 and we need a fix.
The proposed fix doesn't change the semantics, it just explicitly
states that the rule is an expression.
Signed-off-by: Yasushi SHOJI <yashi@spacecubics.com>
The script already accepts a depth parameter to configure the output,
but that parameter was not used at all. This commit adds the correct
handling by passing it to anytrees RenderTree iterator.
Signed-off-by: Detlev Zundel <dzu@member.fsf.org>
Add arm fvp emulator in order to use ninja run or west build -t run.
Add armfvp in order to run twister.
Set env ARMFVP_BIN_PATH before using it,
e.g. export ARMFVP_BIN_PATH=<path/to/fvp/dir>
NOTE: ARMFVP_BIN_PATH is the dir path.
Signed-off-by: Jaxson Han <jaxson.han@arm.com>
When printing the unassigned values the 'sym' variable is
used as a dict from which we try to get the 'name' value.
However, 'symbols['unassigned']' gives a list of keys, so
we get an 'TypeError' when trying to access ['name'] of
a string (the key).
Fix this issue by iterating over the values from
the 'symbols['unassigned']' dict instead.
Signed-off-by: Håkon Øye Amundsen <haakon.amundsen@nordicsemi.no>
Adds feature allowing to use yaml file with dictionaries defining
tests to be quarantined (extra arg "--quarantine-list FILENAME").
The dictionaries are validated according to the proper schema
and loaded.
A flat list is created containing quarantined configurations
(configuration = platform + scenario). Configurations under quarantine
are skipped and get "Quarantine" as a reason in the results reports.
A "comment" can be added to a quarantine entry in the quarantine yaml
with more details (e.g. issue #) and it will be also added to
the report.
The status of tests under quarantine can be verify if
`--quarantine-verify` is used in addition to
"--quarantine-list FILENAME". Using these args will make twister skip
all tests which are not on the quarantine list.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
In CI we typically build for "default" platforms. However some
testcases that set "platform_allow" may have no overlap with the default
platforms and thus thoses testcases will not get built.
Change the platform selection logic in these cases to set the platforms
to the list of allowed platforms ("platform_allow") in the testcase.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
The commit fixes problem with nrfjprog runner for west flash,
that has been introduced with latest changes to BuildConfiguration
class.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The ImgtoolSigner.sign() is passed a BuildConfiguration object instance
that holds all of Kconfig options that have been defined for a build;
yet it has been reading the .config file once again, with the
load_dot_config, to check some of these options.
The commit replaces the code that has been using the load_dot_config
obtained data with the code that uses the BuildConfiguration object.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The body of load_dot_config method has been reimplemented in
BuildConfiguration.get(), replacing the previous code.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The BuildConfiguration.path attribute has been added that stores path
to .config file which has been used as source for options collected to
BuildCOnfiguration object.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
docutils is a core dependency of Sphinx. It is only used by some of our
custom Sphinx extensions.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Add the nanopb library and generator tools as a module.
Nanopb is a small code-size Protocol Buffers implementation in ansi C.
It is especially suitable for use in microcontrollers,
but fits any memory restricted system.
Nanopb home: https://jpa.kapsi.fi/nanopb/
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
This scripts receives the same parameter of what_changed.py. And run
coccinelle scripts for code guideline compliance in the given git
commits. e.g: ./guideline_check.py --commits origin/master..HEAD
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
This coccinelle script can check some violations for rule 21.2.
Currently it is checking the follow reserved names:
"remove", "rewind", "malloc", "free", "exp", "signal"
It can easily be extended in the future though.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Rebnase
This coccinelle script can check some violations for rule 5.7.
It can identify things like:
struct device *device
But it is not capable to identify:
struct test { ... }
...
int test;
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Allow versions starting from 0.5.2 up to < 1.0, be the next major
releases. Major releases tend to include breaking changes, so better
restrict such releases until they are manually tested for potential
regressions.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
We have recently been hit by an issue between docutils and
sphinx_rtd_theme. The problem is fixed since release 0.5.2. In order to
avoid similar problems in the future, pin the theme version.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Add the ability to filter which properties get imported when we do an
include. We add a new YAML form for this:
include:
- name: other.yaml
property-blocklist:
- prop-to-block
or
include:
- name: other.yaml
property-allowlist:
- prop-to-allow
These lists can intermix simple file names with maps, like:
include:
- foo.yaml
- name: bar.yaml
property-allowlist:
- prop-to-allow
And you can filter from child bindings like this:
include:
- name: bar.yaml
child-binding:
property-allowlist:
- child-prop-to-allow
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Replace current filter setup with the warnings_filter Sphinx extension.
Note that current regexes have been simplified to make them more
readable while keeping the most relevant information.
Sphinx warnings are now treated as errors (-W).
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Fixes: #33212
Upstream PR: https://github.com/ulfalizer/Kconfiglib/pull/103
Ignoring when user inputs NULL in a text field.
menuconfig exits with a python stack trace if NULL is provided as input
character, therefore ignore NULL as an input character to prevent this
behaviour.
A NULL character may be given accidentally by the user through the
following ways:
- Pressing `Win` key on keyboard (Windows only)
- Pressing `<CTRL>-@` / `<CTRL>-2`.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
PR #33746 introduced changes to the devicetree python file
that requires changes in the python code that imports the
devicetree module.
This was omitted in the west sign command implementation.
Signed-off-by: Mikkel Jakobsen <mikkel.aunsbjerg@prevas.dk>
We are now in the process of extracting edtlib and dtlib into a
standalone source code library that we intend to share with other
projects.
Links related to the work making this standalone:
https://pypi.org/project/devicetree/https://python-devicetree.readthedocs.io/en/latest/https://github.com/zephyrproject-rtos/python-devicetree
This standalone repo includes the same features as what we have in
Zephyr, but in its own 'devicetree' python package with PyPI
integration, etc.
To avoid making this a hard fork, move the code that's being made
standalone around in Zephyr into a new scripts/dts/python-devicetree
subdirectory, and handle the package and sys.path changes in the
various places in the tree that use it.
From now on, it will be possible to update the standalone repository
by just recursively copying scripts/dts/python-devicetree's contents
into it and committing the results.
This is an interim step; do NOT 'pip install devicetree' yet.
The code in the zephyr repository is still the canonical location.
(In the long term, people will get the devicetree package from PyPI
just like they do the 'yaml' package today, but that won't happen for
the foreseeable future.)
This commit is purely intended to avoid a hard fork for the standalone
code, and no functional changes besides the package structure and
location of the code itself are expected.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Remove dead code from twisterlib.py which is a copy of code
from twister script. This redundancy causes pylint errors in
compliance checks.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
The jlink runner performs a version check which is skipped on
Windows. If running inside WSL we also need to skip.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
Allow a custom (remote) gdb host passed as a debug argument
which tries to connect to an existing one instead of creating one.
This also allows to run the gdb server outside of a WSL
in Windows and the debugger inside of the WSL environment.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
checkpatch requires a Makefile to be present at the top-level directory.
Remove this requirement.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Whenever a child-binding: dict has a compatible, we ought to make
that available in EDT._compat2binding. If we don't, we can't look it
up later.
This has adverse effects like missing child bindings which describe
buses.
Fixes: #32071
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Currently the python script is assuming we only have objects on 32-bit
addresses. This is obviously wrong for 64-bit platforms. Fix this.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
The nrfjprog utility is not capable of flashing a hex file which
affects the flash memories of both coprocessors of the nRF53 family of
SoCs.
However, the user is capable of creating such a hex file using the
HEX_FILES_TO_MERGE build system variable.
An example use case is to build a bluetooth controller application for
the network core, then use the zephyr.hex file in that build directory
as the HEX_FILES_TO_MERGE argument for a separate Bluetooth
application build targeting the app core.
Work around this by detecting the situation and doing the right thing
by splitting the hex file back up again, even if thats a bit awkward.
Splitting the hex into app and network core components allows them to
be flashed separately. This is the only way we can get the job done
with nrfjprog.
This is arguably nicer since there's just one 'west flash' invocation.
At least in the use case named above, you wouldn't need to rebuild the
controller application very often, so this is a simpler user workflow.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
The exported structures that were originally introduced for OpenOCD have
since then been reused for other debugger plugins, including PyOCD and
Segger J-Link.
Rename the Kconfig option and the implementation from openocd to debug
thread info, so that it reflects the fact that this is no longer
specifically tied to OpenOCD.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
The function `handle()` from class `Test(Harness)` had most of its
code being the direct copy of the code from
`Harness.process_test(line)`. This patch replaces the copied code
with the call for `process_test(line)` and removes repetition of
this function call.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
We need all available runners to be defined as subclasses of
runners.core.ZephyrBinaryRunner in order to be able to look them up by
name at runtime. We do this by importing them from runners.__init__.
This process periodically fails when some runner or other cannot be
imported, usually because it is trying to import something outside of
stdlib and not handling ImportError.
Rather than letting this bring down the entire Python process, catch
and log the error. Sort the list again while we're here.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Runners may not import anything outside of stdlib unless they handle
the possible ImportError.
Apply this rule to the bossac runner. The general pattern is to have a
MISSING_REQUIREMENTS global that we handle at the beginning of
do_run(), but in this case this isn't a missing pip package, but a
misconfiguration. Just log a warning; it won't happen in practice.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Runners can (and many should) depend on the devicetree to decide what
to do, especially as it relates to the flash layout. Make it so that
they do not have to manipulate sys.path to get a hold of the edtlib
directory by doing it in one place, before importing any of them.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Add support for including a board-specific config.yaml file
in the pyocd flash command. Similary to openOCD, the config
file is placed in the board directory under support/.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Use edt.label2node in dt_compat_enabled_with_label DT filter.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Co-authored-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
The patch improves handling runners errors from flashing. It
add thread termination in case of process returning non 0 exit
code. Before, even though the error was returned,
a thread was not terminated and test ended
only after timeout termination. The patch also adds information about
the failure reason to the reports
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Update test to account for new default parameters that allow
use image smaller than flash device capacity.
Signed-off-by: Jose Alberto Meza <jose.a.meza.arellano@intel.com>
The function returns true if the node with the referenced
node label is compatible and enabled.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
The script gen_handles.py was introduce in #32127 but relies on
ZEPHYR_BASE being set in environment.
However, it is not a requirement to set Zephyr base in the users
environment, therefore this is changed to an argument `-z` or
`--zephyr-base` which will be passed from the build system to the
script.
If `-z` or `--zephyr-base` is not provided, the environment will be
checked for a ZEPHYR_BASE setting there.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
BabbleSim execution was (AFAICT mistakenly) disabled in
dda6a5ee90d0f5b31e7d269761dc2ab1a2f8510c. Re-enable it so we run them in
CI again.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Add execution time for testing result of each ztest testcase as:
START - test_sem_multi_take_timeout_diff_sem
PASS - test_sem_multi_take_timeout_diff_sem in 2.54 seconds
Fix#32137.
Signed-off-by: Enjia Mai <enjiax.mai@intel.com>
Logic for selecting non-default platforms listed in
integration_platforms did not work. Functionality is restored in this
change.
Fixes#32835
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Commit 0d4dca10b2 ("scripts: edtlib:
child binding compatibles match parents") was a hack meant to keep the
edtlib.Binding class in place without modifying some twister behavior
that needed further changes to work properly with first-class binding
objects.
This is a hack and is no longer necessary, so back out of this change.
Child Binding objects now have None compatible properties unless the
binding YAML explicitly sets a compatible.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Originally added in 7733b94224.
This filter is not well-formed. It's meant to match nodes like
/leds/led_0 in this DTS:
/ {
aliases {
led0 = &led0;
};
leds {
compatible = "gpio-leds";
led0: led_0 {
gpios = <...>;
label = "LED 0";
};
};
};
Uses look like this:
filter: dt_compat_enabled_with_alias("gpio-leds", "led0")
But notice how the led_0 node doesn't have compatible "gpio-leds";
it's actually the *parent* node that has that compatible.
Replace this with a new filter, dt_enabled_alias_with_parent_compat(),
which is used like this:
filter: dt_enabled_alias_with_parent_compat("led0", "gpio-leds")
This has a name and argument order that makes the meaning of the
filter clearer.
Replace in-tree users with the new filter.
Deprecate the old filter and warn about its use using the standard
logging module.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
