Testcase execution time doesn't match between twister.xml and
twister.log. Testcase execution time is the sum of the previous
testcases' execution time plus its own execution time in
twister.xml.
This patch fixes the issue above.
Signed-off-by: Ioannis Damigos <ioannis.damigos.uj@renesas.com>
Provide a mechanism to propagate useful arguments from one runner to the
next. The primary use case for this is to propagate a JLink serial
number, so that if it is queried from the terminal the user only needs
to make the choice once.
Implements #76077.
Signed-off-by: Jordan Yates <jordan@embeint.com>
The recently introduced board.full_name property can now be used as part
of the format string in the `west boards -f ...` command.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
For pristine builds 'west build' will now create a build_info.yml file
containing the west build command including arguments.
This is done to help users and external tools to recreate builds.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
The build_info function provides a generic and stable way of dumping
build information to the <build>/build_info.yml file.
The build info file is in YAML format and the keys in the file are
intended to be stable, as to allow external tools to retrieve
information regarding the build.
The main differences to the CMakeCache.txt are:
- Settings in the CMakeCache.txt are user controlled, whereas the
information in the build info file is intended to be those values
which are used by the build system regardless if those are specified
by the developer or picked up automatically.
- Internal build system variables are not present in the CMake cache
and should not be, because their values are calculated when CMake
runs.
This also has the benefits of decoupling CMake variable names from
build info keys. Several CMake variables has internal build system
names, and the build system is free to rename those at its own
discretion.
Having dedicated key names ensures a stable API that external tools can
rely upon.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
When building an LLEXT-enabled kernel, 62b19ef65c added weak aliases
of all syscall implementation functions to a pointer to NULL, with the
assumption that LLEXT would check the required symbols at link time and
fail if any of them were found.
This check, however, is ineffective in the current implementation: the
actual address that is exported is the rather normal-looking location of
the variable containing the NULL pointer. This defeats the NULL symbol
validity checks in llext_link.c and causes the extension to crash at
runtime by jumping to a location containing a few zeroes in read-only
data memory.
This commit makes sure the alias target is actually placed at address 0
using the llext-sections.ld linker fragment, so that undefined syscall
implementations are exported as NULLs and as such properly flagged at
link time.
The test for this functionality is also updated to reflect the change.
Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
Full name or description of a board is something we are missing in
HWVv2. It is right now being added to yaml files parsed by twister. This
should be generically available to tooling and documentation
independently from twister.
As we rework how twister parses board meta-data (#77250) and how we
generate board documentation (#79160), this becomes neceassry.
Moving the board full name/description from the twister yaml files to
the board.yaml is something we can automate once the schema is agreed
upon.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Like some other string properties, I will add a derived form
to FULL_NAME to make it easier to reference from macros.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
This is required if boards make use of such definition in their Kconfig
files. In Kconfig.board.v2, only `boards/Kconfig.v2` is loaded, but
BOARD_REVISION is part of `boards/Kconfig`, which can't be loaded in this
context.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Fixed escaping of double quotes, backslashes, and new line characters
so they can be used in string properties.
Previously, double quotes and backslashes were escaped in gen_defines.py
but not in gen_dts_cmake.py, and new lines were not escaped in either,
so using any of these characters would break the build.
Signed-off-by: Joel Spadin <joelspadin@gmail.com>
Add unsigned integer support to the log parser.
This does not change the underlying log format,
it only allows the log parser to more accurately
read the log format.
Signed-off-by: Georges Oates_Larsen <georges.larsen@nordicsemi.no>
Fix trailing `\\r\\n` (escaped CR/LF) didn't cut off because of rstrip()
removed by #58338, so the CR/LF suffix was never found as the actual line
end was `\\r\\n\n`.
Add ANSI code sequence to `test_handlers` Twister unit test.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Add several examples for `upload_test_results_es.py` script usage
with ElasticSearch index map files for the following use cases:
* Twister test results.
* Twister test results with recordings.
* Memory Footprint data (`twister-footprint.json`).
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Separate the pickled EDT generation from the C-Macro header
generation in gen_defines.py to have a more clear responsibility
of the scripts in the DTS parsing process.
Signed-off-by: Benedikt Schmidt <benedikt.schmidt@embedded-solutions.at>
Introduces type hints to all functions for improved static type checking
and IDE support.
Also equalizes spacing between functions as the lines are being touched
anyway.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
This commit adds support to dictionary logging to parse binary
data directly from input serial and generate the ascii logs.
Signed-off-by: Kumar Siddaramayya <harish.kumar@nordicsemi.no>
native_posix is now deprecated.
Building this sample in both native_sim and native_posix does not
improve coverage for the sample or subsystem but doubles CI time.
As anyhow native_posix will be removed all together in
2 releases, let's remove it already for this sample.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Adds a --download-buffer-size option to the canopen runner to allow
specifying the buffer size of the SDO download.
Signed-off-by: Fabian Blatz <fabianblatz@gmail.com>
Package in the requirements-run-test.txt file, bz,
is not package enabling bz2 support.
It is a security concern and must be removed.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Currently, debug logging in the console and verbosity
are tightly coupled - verbosity of level 2 and higher
enables logging at the debug level.
This change introduces a separate Twister flag
responsible for controlling the debug logging,
while leaving the rest of verbosity unchanged.
This allows for controlling the verbosity on
both logging levels, according to one's needs.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
In case of problem with parsing hex data from coverage dump,
do not create empty gcda file.
Such empty file will break gcovr parsing.
Signed-off-by: Piotr Kosycarz <piotr.kosycarz@nordicsemi.no>
Added new Kconfigs for defining permission level of GATT
characteristics that are part of the Bluetooth SMP service in the
MCUmgr subsys.
Removed the CONFIG_MCUMGR_TRANSPORT_BT_AUTHEN as the new Kconfig
options are mutually exclusive and need to be groupa as the Kconfig
choice option.
Signed-off-by: Kamil Piszczek <Kamil.Piszczek@nordicsemi.no>
Add new script `pack_as_twister.py` to convert memory footprint
data prepared by `./footprint/scripts/track.py` into JSON files
compatible with Twister report schema. Next, the data can be
transformed and uploaded to ElasticSearch data storage the same
way as memory footprint (and other) reports executed by Twister.
Add to `plan.txt` an optional column with the corresponding
test suite names for 'footprints' as an example for test instance
name composing with `--test-name` command argumnent.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
This small change concerns the following filter functions:
1. `dt_compat_enabled(C)`:
There's a node with compatible `C` and status "okay".
2. `dt_enabled_alias_with_parent_compat(A, C)`:
There's a node with alias `A` and status "okay", and its parent
has compatible `C`.
3. `dt_label_with_parent_compat_enabled(L, C)`:
There's a node with label `L`, and its parent has compatible `C`
and status "okay".
All three functions involve checking whether some node or its parent has
a given compatible, but the way this has been checked is inconsistent.
Function (1) has done it with this Python conditional:
compat in node.compats
while (2) and (3) have used:
parent.matching_compat == compat
The first check works well with nodes that have multiple compatibles,
and it is more aligned with the notion of "has_compat" as seen in the
devicetree macros for C, CMake, and Kconfig.
Arguably, `matching_compat` shouldn't have been used here, because it is
actually a property of a node's binding, moreso than of the node itself.
In practice, it's usually equal to the first compatible for which edtlib
has found a binding, which at first glance is just more constrained than
the `node.compats` check. However, there also exist obscure cases where
the `node.compats` are empty, while the `node.matching_compat` is not.
For now, the three functions can use a combined check, to improve
consistency and utility while avoiding breakage:
node.matching_compat == compat or compat in node.compats
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
It has been deprecated since Zephyr v2.6.0.
The filter that replaced it - `dt_enabled_alias_with_parent_compat` -
had shared code with it, which can now be inlined into `ast_expr()` to
match the parser's overall code structure.
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
The script was only looking at the Zephyr base repository and failing to
look for soc roots e.g. in other modules.
Signed-off-by: Johan Hedberg <johan.hedberg@silabs.com>
With the requirement in #78316 for Zephyr modules to always define
ZEPHYR_<name>_MODULE Kconfig setting then there is no longer a need
for this commit. Simplify check_compliance by reverting ths commit.
This reverts commit 35e28e6315.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
When at verbosity 1, we print out the status of TestInstances.
This makes it harder to notice changes at TestCase level,
which require perusing the logs.
This adds TestCase status and reason printing
if verbosity level is 2 or more.
Reason printing is suppressed if the reason is empty or None.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Running sysbuilt tests fails because of missing "run" target.
This adds the default domain context to the command.
Signed-off-by: Wilfried Chauveau <wilfried.chauveau@arm.com>
Some dict.get() calls did not use a TwisterStatus
as a default value, thus using a NoneType where
TwisterStatus should appear.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
ReStructuredText can sometimes be tricky to get right, especially for
folks that might be more familiar with Markdown.
This adds a Sphinx/RST linter to the compliance check script to help
catch common issues that can easily go unnoticed and cause rendering
issues.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Fixes bad usage of single backticks in lieu of double backticks for
rendering inline literals, or simple '*' for italics.
When appropriate, a better construct than double backticks has been
selected (ex. :file:, :kconfig:option:, :c:func:, ...), or proper :ref:
have been used if the original intention was to have a link.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
If you interrupt process() operation, we want Twister
to exit as gracefully as it can. This avoids the
UnboundLocalError that could appear e.g.
when interrupting the operation via SIGINT.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
"type | None" syntax is only available with Python from version 3.10,
fix building with earlier 3.x Python versions.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Align with native_simulator's upstream main
b4b9791ac822ae300363dc3ebbc7b7ac200632ce
Which includes:
* b4b9791 NATIVE_SIMULATOR_IF macros: Provide version with explicit
section names
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
This changes how some arguments are set in the `Handler`s.
`options`, `generator_cmd` and `suite_name_check` are now passed as
arguments to the constructor rather than injected from an other module.
Signed-off-by: Wilfried Chauveau <wilfried.chauveau@arm.com>
Python 3.12 warns that
> Testing an element's truth value will raise an exception in future
> versions. Use specific 'len(elem)' or 'elem is not None' test instead.
> if elem_ts := root.find('testsuite'):
Signed-off-by: Wilfried Chauveau <wilfried.chauveau@arm.com>
CONF_FILE, DTC_OVERLAY_FILE and OVERLAY_CONFIG are deprecated but still
used by the tests causing warnings when running them.
This adds a test_data specific to validate the emission of the warning,
and removes the offending args from the other test_data files.
Signed-off-by: Wilfried Chauveau <wilfried.chauveau@arm.com>
Space separated lists are deprecated but this notice is not checked for.
extract_fields_from_arg_list also converts lists back to space-separated
lists causing a warning on get_scenario
Signed-off-by: Wilfried Chauveau <wilfried.chauveau@arm.com>
Moves the telnet client into runners/core.py as well, as this is now shared
between openocd and jlink.
Signed-off-by: Tobias Pisani <mail@topisani.dev>
This was non-trivial, as openocd is a bit weird to work with. Using only
commands passed with '-c' arguments, I couldn't get it to reliably resume
(or just not halt) the target when started. I tried using the 'sleep'
command, and various 'configure -event XX { resume }' events, but nothing
panned out, as it seems to always halt after all `-c` commands have been
run.
To avoid that, this waits for the TCL RPC port to be up, and sends a
resume command there. This works reliably.
Signed-off-by: Tobias Pisani <mail@topisani.dev>
This command runs separately from a debug server, instead of attaching
to a running server. This is both the easiest out of the box experience,
and also should be possible to implement consistently for most runners.
This commit includes an initial implementation for pyocd.
Signed-off-by: Tobias Pisani <mail@topisani.dev>
Multiple improvements of the `upload_test_results_es.py` script:
* JSON objects flattening.
This feature allows `twister.json` file preprocessing to simplify
its Elasticsearch index structure for complex hierarhical objects,
for example with memory footprint, or code coverage data.
A new command line option `--flatten` is added to change testsuite data
structure in regard of one of its list objects: either `testcases` or
`recording`, so each item there becomes an independent data record
inheriting all other testsuite properties, whereas the children
object's properties are renamed with the parent object's name
as a prefix: 'testcases_' or 'recording_' respectively.
Only one testsuite property can be flattened this way per index upload.
Other children objects will be treated accorging to the index structure.
Related new command line options (with help text explanations):
`--flatten-dict-name`,
`--flatten-list-names`,
`--flatten-separator`,
`--transpose-separator`,
`--escape-separator`
* A new command line option `--transform` is added to allow regexp group
parsing in string propertites extracting additional derived properties.
* A new command line option `--exclude` is added to exclude testsuite
properties not needed to store at Elasticsearch index.
* Branch name `--run-branch` and Workflow ID `--run-workflow` command
line options as additional key fields to allow data from different
branches, workflows and triggering events in the same index.
* A new command line option `--map-file` is added to apply
an explicit index structure to the `twister.json` input data.
* Add bulk operation timeout parameter for heavy/long uploads.
Other changes:
* batch upload error handling and logging;
* inline documentation improvements;
* some corner case fixes on empty objects.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Status errors previously logged an error, but didn't fail the running test.
This commit changes that
and introduces a new StatusAttributeError to use there.
One test is modified so it follows proper status form.
One test for the new error has been added.
Status errors now will properly mark the Instance as ERROR
and not run TestCases as SKIP.
This necessitated some code layout changes in runner.py
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
This patch adds bindesc support for the build version values for the
kernel and application - BUILD_VERSION and APP_BUILD_VERSION.
The kernel's BUILD_VERSION can be overridden at build time.
Signed-off-by: Attie Grande <attie.grande@argentum-systems.co.uk>
Get data for various roots from modules and do not hardcode location
when retrieving list of boards in testplan.
Fixes#71761
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
For any module that defines blobs, add a new Kconfig symbol to indicate
whether the blobs have been fetched or not. Example output for the
hal_silabs module:
# (no blobs present)
$ scripts/zephyr_module.py --kconfig-out=/dev/stdout \
-m ../modules/hal/silabs
menu "hal_silabs (../modules/hal/silabs)"
osource "/Users/johedber/src/zephyr/modules/hal/silabs/zephyr/Kconfig"
config ZEPHYR_HAL_SILABS_MODULE
bool
default y
config ZEPHYR_HAL_SILABS_MODULE_BLOBS
bool
endmenu
$ west blobs fetch hal_silabs
# (blob fetching output)
$ scripts/zephyr_module.py --kconfig-out=/dev/stdout \
-m ../modules/hal/silabs
menu "hal_silabs (../modules/hal/silabs)"
osource "/Users/johedber/src/zephyr/modules/hal/silabs/zephyr/Kconfig"
config ZEPHYR_HAL_SILABS_MODULE
bool
default y
select TAINT_BLOBS
config ZEPHYR_HAL_SILABS_MODULE_BLOBS
bool
default y
endmenu
The generated output for modules which do not define blobs is not affected.
Having this additional symbol for blobs lets us specify Kconfig
dependencies for features which require the blobs to be present.
Signed-off-by: Johan Hedberg <johan.hedberg@silabs.com>
As the .bin & .hex build output is optional
and it can be disabled by CONFIG_BUILD_OUTPUT_BIN/HEX,
add support for the mandatory .elf build output
to the linkserver runner flash command.
Signed-off-by: Andrej Butok <andrey.butok@nxp.com>
If a diff only has added or removed lines we need to match both '-' and '+'
characters in the hunk context.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
Re-arrange Twister command line options for test plan reporting
to a dedicated group with mutally-exclusive options to reflect
actual implementation and its dry-run execution mode.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Re-run issue check when a PR is updated, i.e. when someone adds
'Fixes...` to the PR body.
This is mostly for release branches and has no effect on main branch.
Also, add concurrency check in the workflow.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Update check_compliance to handle auto defined ZEPHYR_<name>_MODULE
Kconfig symbols.
check_compliance runs three levels of Kconfig check.
A basic and a full, both which uses the generated Kconfig.modules
created according to Zephyr modules present.
A Kconfig check where no Zephyr modules are sourced. This check ensures
that Zephyr Kconfig tree doesn't refer to Kconfig symbols defined in
Zephyr module's local Kconfig trees.
However, there are a few auto generated symbols which are allowed,
such as: `ZEPHYR_<name>_MODULE` and `ZEPHYR_<name>_MODULE_BLOBS`.
Therefore, when testing no blobs, filter the generated Kconfig.modules
file, so that no sourcing of extra Kconfig files are performed but
lines defining `ZEPHYR_<name>_MODULE` and `ZEPHYR_<name>_MODULE_BLOBS`
are preserved.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
In RISCV "sepc" is actually the name of an instruction.
As this prevents building. We should remove the line.
Signed-off-by: Sven Ginka <s.ginka@sensry.de>
The clang-format check outputs a git diff with surrounding context. It
naively removed the first and last 3 lines, but this does not work if
there are less lines.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
As the .bin & .hex build output is optional
and it can be disabled by CONFIG_BUILD_OUTPUT_BIN/HEX,
add support for the mandatory .elf build output
to the pyocd runner flash command.
Signed-off-by: Andrej Butok <andrey.butok@nxp.com>
To specify the serial number, JLink expects either one argument
('--dev-id=xxx') or two (e.g. '--dev-id' 'xxx'), but it can not deal
with a single one that is '--dev-id xxx'.
The problem has been introduced (or just made visible?) by commit
5ee4284320 (twister: runner: j-link: use
dev-id instead of SelectEmuBySN) in PR #76931.
How to reproduce:
1. Create a HW configuration map, e.g.:
```
$ cat zephyr-hw-map-nrf52840dk-1.yml
- connected: true
id: '683517317'
platform: nrf52840dk/nrf52840
product: nRF52840 DK 1
runner: jlink
serial: /dev/ttyACM-nrf-dk-1
```
2. Run test `logging.dictionary` with Twister:
```
$ west twister --platform nrf52840dk/nrf52840 --device-testing \
--hardware-map zephyr-hw-map-nrf52840dk-1.yml -s logging.dictionary
```
3. The build will fail, and the `twister_harness.log` contains:
```
10:21:24.375:DEBUG:twister_harness.device.factory: Get device type
"hardware"
10:21:24.375:DEBUG:twister_harness.device.hardware_adapter: Opening
serial connection for /dev/ttyACM-nrf-dk-1
10:21:24.376:DEBUG:twister_harness.device.hardware_adapter: Flashing
device 683517317
10:21:24.376:DEBUG:twister_harness.device.hardware_adapter: Flashing
command: <snip>/bin/west flash --skip-rebuild --build-dir
twister-out/<snip>/tests/subsys/logging/dictionary/logging.dictionary
--runner jlink '--dev-id 683517317'
10:21:24.590:ERROR:twister_harness.device.hardware_adapter: Could not
flash device 683517317
10:21:24.592:DEBUG:twister_harness.device.hardware_adapter: Closed
serial connection for /dev/ttyACM-nrf-dk-1
```
(note the '--dev-id 683517317' part)
4. Running the stated `west flash` command shows the following error
message:
```
-- west flash: using runner jlink
FATAL ERROR: runner jlink received unknown arguments: ['--dev-id
683517317']
```
Signed-off-by: Reto Schneider <reto.schneider@husqvarnagroup.com>
Check if lines from serial are processed one by one
in case when was received in one buffer from redline method.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
Readline method sometimes receives more lines in buffer.
Split them to avoid misinterpreting data in harness module.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
Use pytest-args parameters provided in command line after
parameters taken from yaml file. The last occurence is
considered by argparse, so parameters from command line
will be used.
Fixes: #77319
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
This will update the posix thread names to match
the zephyr thread names.
This will simplify debugging as the debugger will
recognize the thread names.
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>
The gtest 1.15 has slightly different output from the original one that
was used to set up the harness. With this change we support both output
formats.
Fixes#72318
Signed-off-by: Yuval Peress <peress@google.com>
errno is defined by the ISO C standard to be a modifiable lvalue of type
int, and **must not be explicitly declared**
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Do not try to format each single failure, this will take too long to
process and will run for hours and spam GH.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Modify the condition in the parse_testplan
function to align with the changes in the Status class.
Signed-off-by: Arkadiusz Cholewinski <arkadiuszx.cholewinski@intel.com>
Add minimum support for the `west sdk` command.
This only provides completion for the `sdk` command itself.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
This command can list and install SDK.
Run 'west sdk install' to install the SDK.
Run without any parameter, installing
specified by SDK_VERSION in the source tree.
'west sdk' to show installed SDK information.
This command is just a wrapper for SDK's setup command,
but it simplifies the installation process. It will be a good
improvement for onboarding first-time users.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
The .bin & .hex build output is optional and can be disabled by
CONFIG_BUILD_OUTPUT_BIN/HEX.
Add support for the mandatory .elf build output
to the jlink runner flash command.
Signed-off-by: Andrej Butok <andrey.butok@nxp.com>
The serial number for debugger selection over USB
can be selected with the dev-id. This change
reflects also more the workflow of west flash
with J-Link.
The usage of SelectEmuBySN breaks the support for
J-Link over IP with twister.
Signed-off-by: Michael Arnold <marnold@baumer.com>
For regexes such as `CONFIG_DEBUG_(OPTIMIZATIONS|INFO)` and
to be able to use `.*` as this syntax is used in the Kconfig search.
Signed-off-by: Tomi Fontanilles <tomi.fontanilles@nordicsemi.no>
Extend pytest-args from configuration yaml file with args
from command line instead of overwriting.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
In order to reduce excessive false positives for Coding
Guidelines checks in CI, maintain a list of paths that are
marked "safe" for reserved names to be implemented or declared.
Signed-off-by: Chris Friedt <cfriedt@tenstorrent.com>
Fix confusing `TypeError: 'type' object is not subscriptable`
at _get_installed_packages() if twister runs on python version < 3.9
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Improve DUT selection at DeviceHandler: for each DUT it counts
how many test instances have been failed on it during the current
twister execution, so the next available DUT will be chosen
ordering the eligible DUTs by less failures occured so far.
The new selection mechanism should increase chances to retry failed
tests on different DUTs, for instance to resolve ploblems when some
DUTs have connectivity or HW issues slowing down test plan execution,
or even block the execution when only one test suite runs whereas
the same first DUT candidate in the list is not working and others
were not chosen.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Change Twister PyTest plugin's test finalizing sequence to release
the DUT it is used as the very last operation, after the Test Instance
status becomes fully updated from the execution results.
This also fix a race condition possible when pytest plugin releases
the DUT and it becomes acquired by another test while the current test
is not yet finalized completely.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Twister DeviceHandler - add test failure counter for how many
test instances have been failed on each DUT (Device Under Test)
when it executes the current test plan.
Output DUT falure counter summary at the end of Twister run.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Fix Twister DeviceHandler exit on SerialException when
it connects to the serial device in 'flash before' mode.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Several improvements at Twister DeviceHandler when it releases
current DUT (Device Under Test):
- release the exact DUT which is used for the test instance
instead of all configured DUTs which happened to have
the same serial device configured.
- Twister PyTtest harness plugin adjustment to the above.
- additional debug logging to track DUT waiting/retain/release.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Updated the Pytest-Harness methods to capture and log error messages from
the subprocess, when pytest command fails.
Ensured that error messages are logged with `--inline-log` option
and placed in twister log reports.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
While the option to inform pytest about which platform a test is for
(platform being the board, such as mps2/an385) is available, it wasn't
being used. This patch fixes that, by building a pytest command which
includes the platform.
Signed-off-by: Ederson de Souza <ederson.desouza@intel.com>
Add a new fixture, `unlaunched_dut`, which is basically the `dut`
one, but without launching the device - thus also not building the
application. It is useful for tests who need a finer control of
the building of the application, before the dut can be launched
to run it.
It will be used on a future patch, which will use it to enable LLEXT EDK
tests.
Signed-off-by: Ederson de Souza <ederson.desouza@intel.com>
Fix after #71401 merge at Twister Reporting:
* filter status value type mismatch;
* tesplan.json generation incorrect error logs.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Gets experimental and deprecated symbols directly from Kconfig instead
of rely on hardened.csv.
This way we keep the tool consistent with Zephyr's code.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Samples, that were obtained by profiling perf tool, can be be translated
into flamegraph using stackcollapse.py script.
Originally-by: Yonatan Goldschmidt <yon.goldschmidt@gmail.com>
Signed-off-by: Mikhail Kushnerov <m.kushnerov@yadro.com>
Incorrect status assignments were previously just warnings in logs.
Now they are elevated to errors.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Various different Statuses were joined into a single class,
TwisterStatus. This change anticipates further streamlining
of the Twister's approach to Status.
Code guarding Twister's Properties was shortened to a
value check only.
QEMUOutputStatus was left separate, as doubts were cast
whether it should remain a status. Leaving it separate makes
its removal easier.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
TestInstanceStatus of TIMEOUT and FLASH were never really used.
They were checked for, but never assigned.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Now statuses are not just a str that can be easily mistyped
or assigned wrong. Now they are an Enum.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
This change is removing some deprecation warnings
which for some reason causing failing tests with
Python 3.12 on CI #76877.
Also, it fixes warnings from pytest like:
PytestCollectionWarning: cannot collect test class 'TestPlan'
because it has a __init__ constructor
(from: scripts/tests/twister/test_testplan.py)
Signed-off-by: Lukasz Fundakowski <lukasz.fundakowski@nordicsemi.no>
Add a new compliance check that reports any clang-format issues on
the git diff and prints a warning.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
Do not depend on platforms that need a HAL. This should speed things up
and should resolve issues where runner did not have enough space to deal
with all HALs.
t
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Driver for Nordic nRF70 Wi-Fi6 companion chipset, depends on
hal_nordic/nrf_wifi for OS agnostic part of the driver.
This supports (Q)SPI interface to communicate from host to chip.
Signed-off-by: Chaitanya Tata <Chaitanya.Tata@nordicsemi.no>
When presenting an example of how to run a failing test case with
`west build`, provide the source dir path relative to the current
working directory, not the zephyr root directory.
Signed-off-by: Jordan Yates <jordan@embeint.com>
It is safer to base the SUIT artifacts path on the path of the
configuration file than the HEX file that is being flashed.
The latter may be overriden by several scripts that merge/transform the
final firmware image.
Signed-off-by: Tomasz Chyrowicz <tomasz.chyrowicz@nordicsemi.no>
Read CONFIG_SHELL_PROMPT_UART from config file and use them
in shell fixture in pytest-harness package.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
Moved helper methods from tests/boot/with_mcumgr to pytest-harness
package. It can be reused by other tests.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
Add support in arm_cortex_m python script to read thread
registers off of a thread's stack when switching context.
When CONFIG_ARM_STORE_EXC_RETURN is enabled, check the exc_return value
in thread's arch struct to determine accurately where the stack pointer
is. Also, set r7 (the frame pointer register) in case the frame pointer
is not omitted.
Only 8 registers are read from the top of the stack
for other threads present in the dump. So update the
script to reflect that.
Signed-off-by: Mark Holden <mholden@meta.com>
Update zephyr gdb-server scripts to understand threads.
Parse the kernel_thread_info out of the elf file to be used
for finding offsets to data from _kernel structs or from
individual threads.
Update log_parser to understand latest format change, which
allows for the presence of a new section, threads metadata.
Update gdbstub to respond to various packets to describe
the threads present in a dump, and allow switching to
thread context of each thread.
Signed-off-by: Mark Holden <mholden@meta.com>
Twister scans C-files to find testcases that are implemented
using ZTest framework. Also runs scanning of Elf files
after building. Skip scanning files if it is not required.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
probe-rs is a new programming and debugging tool written in Rust, supports
many probes and targets.
This commit introduces initial support for probe-rs to Zephyr.
Signed-off-by: Chen Xingyu <hi@xingrz.me>
Running 'west blobs fetch' does not verify the digest of downloaded files:
1. if the checksum of the previously downloaded file does match
that in the blob metadata (status BLOB_PRESENT), do nothing
2. if the checksum of the previously downloaded file does not match
that in the blob metadata (status BLOB_OUTDATED),
download the "up to date" file
3. if the blob has not yet been downloaded (status BLOB_NOT_PRESENT),
download it
None of the 2) and 3) code paths will verify that the checksum of the file
just downloaded actually matches the digest in the blob's metadata.
In the event that the metadata of a module is incorrect, then the user
will not notice anything, and may rely on an unexpected binary,
e.g. a static library for a different architecture.
According to the Binary Blobs documentation [1], the expected
behavior is to check the blob digest after downloading.
[1] Fetching blobs, Zephyr 3.6.0 (still applies to Zephyr 3.7.0rc3)
docs.zephyrproject.org/3.6.0/contribute/bin_blobs.html#fetching-blobs
Signed-off-by: Christophe Dufaza <chris@openmarl.org>
With the recent change to hwmv2, loading the
boards is extremely slow and can take several
seconds.
To solve that problem this commit add a cache
of the boards. The cache is updated based on:
- when the latest commit of the manifest
file directory is updated or;
- when the manifest file directory is not
a git directory, when the manifest file
content itself is updated.
At the same time:
- update how the board completion is
displayed by including the board vendor to it;
- add missing `--board` and `--board-dir`
options;
- remove `hwmv1` board completion code.
Signed-off-by: Théo Battrel <theo.util@protonmail.ch>
The current option used as an example, --recover, is actually a separate
option in the runner itself. Instead use --clockspeed as an example,
which is applicable to all nrfjprog commands.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Allow for users to provide a --qspiini parameter that is passed directly to
the nrfjprog executable but only in the --program operation. This is
required since e073210ec2 enabled the
-O/--tool-opt for all operations, but --qspiini is only allowed combined
with --program.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
When --erase was specified, esp32 runner was autodetecting serial port to
be used, regardless of --esp-device argument.
Append '--port SERIAL_DEVICE' parameter earlier, so that erase command
invocation uses explicitly specified serial device.
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
The default base timeout for pytest is statically set by
the TwisterHarnessConfig class to be 60 seconds. However,
sometimes it takes longer than 60s before the app starts
to run, especially on emulator/simulator where it takes
quite some time to start. So pass the test timeout as
the base timeout via pytest command line argument.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Current implementation will not work if comits were not provided.
ie. use case with list of changed files will fail as args.commits is None.
Signed-off-by: Piotr Kosycarz <piotr.kosycarz@nordicsemi.no>
Doing duplicates count of bugs, a PR fixing a bug is not a bug report.
Many PRs fixing an open bug are labeled with 'bug' and thuse are being
counted twice.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Address the cases where submitter is also the maintainer of the code
changed and other areas are being changed. In this case, assign to the
next area maintainers instead of assigning to submitter.
Example: maintainer of component A introduced significant changes to
area A but also makes changes to other areas B and C. Right now
maintainers of B and C are added as reviewers.
This change will assign to the next area after A, i.e. B in cases where the
submitter is also the maintainer of area A.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
When validating the flash runner configurations in `soc.yml`, the only
SoCs that were considered had to be defined under this structure:
family:
- series:
- socs:
- name: ...
However, the `family` and `series` keys are optional, so the `soc.yml`
files can also be arranged like this:
family:
- socs:
- name: ...
series:
- socs:
- name: ...
socs:
- name: ...
The solution is to move the validation code further down, so that it can
reuse the SoC data that was already correctly parsed while initializing
a `Systems` instance.
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
Updates the minimum version of imgtool to 2.1.0, which is one year
newer than the 2.0.0 release
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
Previous OpenOCD version finding would fail when additional tokens were
prepended to the 'openocd --version' output, as happens with some third-
party OpenOCD repackages (xPack for one).
Fixes: #71955
Signed-off-by: Nick Kraus <nick@nckraus.com>
Disable branch coverage for the `__ASSERT` family of macros. Covering
all of the assertion branches by definition means triggering the
assertion, which can be either challenging or impossible to exercise,
and in either case results in the immediate termination of the test.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Multiple values for `--exclude-branches-by-pattern` will result in only
the last value taking effect. Resolve this by merging all the provided
regex patterns into a single pattern with the `|` operator.
Signed-off-by: Jordan Yates <jordan@embeint.com>
The minimum version of pyelftools is 0.29 to make it working
with scripts/footprint/size_report
Fixes#75605
Signed-off-by: Sylvio Alves <sylvio.alves@espressif.com>
Fixes#71761
The `west boards` command parses extra BOARD_ROOTs from Zephyr modules,
so that the boards defined in those modules are automatically listed.
In HWMv2, OOT boards can be described in terms of OOT SoCs, which means
that extra SOC_ROOTs must also be provided. Otherwise, an error message
will be displayed when attempting to list all boards. Therefore, every
Zephyr module SOC_ROOT should be included as well.
In HWMv1 (deprecated), OOT boards can be defined in terms of OOT archs,
but module ARCH_ROOTs had never been included automatically. The fix for
this is long overdue, but it's included for symmetry.
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
The gTest harness asssumed that the lines end with the test name, but
some gTest implementations include the test duration in the line. Update
both the tests and regex to allow this and also avoid capturing
characters into the `test_name` that cannot be valid test name chars.
Fixes#72318
Signed-off-by: Yuval Peress <peress@google.com>
Increase the default SDO timeout for the CANopen program download west
runner from 0.3 seconds to 1 second. Depending on the flash size and speed,
a full erase may take slightly longer than 300 ms.
The timeout can be customized by using the --sdo-timeout runner parameter.
Fixes: #73987
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
The current version of scipts do not consider OOT boards use cases and
the tests with robot now are strict to only one robot file, which is
not realistic for real environment. This address those issues and allow
multiple testsuits at command line and lists at tests entries. It add
another test parameter to allow configure robotframework options.
Fixes: #74563
Signed-off-by: Gerson Fernando Budke <gerson.budke@ossystems.com.br>
Ignore changes to the CANopen program download west runner in CI.
This script is only executed when performing a DFU using the CANopen
protocol via CAN, which is never triggered in CI anyways.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
The flag won't work with sysbuild since there is no way to
reliably tell to a parser which data came from which image.
fixes: #74092
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
pylint keeps failing and complaining about arg_data_type is
used before assignment. So assign it to None to silence
the warning.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The package_len has been extended from 10 bits to 11 bits
in the log message header. So the format for dictionary
logging also needs to be updated.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Python does not really support long long double, so %llx cannot
be formatted correctly, so we replace it with a simple %lx.
There is another variant %#llx and we also need replace it to
%#lx.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
... and put them into the LogParser class file instead of
the verisoned parser. This is in preparation for introducing
a new parser version.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Printing long long requires alignment on 64-bit before parsing
the actual argument. Or else the parser would be looking at
some unrelated bits. So fix it.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This extracts the DataTypes class into its own file. This is in
preparation to add a new version of parser which can reuse this
class.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Adjust error message so that it clearly states runners.yaml is
missing from <build_dir>/zephyr, instead of referencing CMake cache
variable ZEPHYR_RUNNERS_YAML, which is no longer used (since
3124c02987 ).
Also clean up that variable in CMake since it is no longer used
(0 other references in entire tree).
Fixes#70605
Signed-off-by: Louis Feller <louis.feller@st.com>
`checkpatch.pl` requires that dts sources are indented with tabs,
fix all the spaces that slipped in while checkpatch wasn't watching.
Signed-off-by: Jordan Yates <jordan@embeint.com>
When working with coredumps, it is useful to be able to modify base
registers. Adding this capability allows implementing scripts to
inspect backtrace of threads other than current the current thread.
Signed-off-by: Félix Turgeon <felixturgeon@meta.com>
In size_report script, if the DWARF section of ELF file contains both
debug_loc and a debug_loclists sections, LocationListsPair class is
used to track locations. In that case, parse_from_attribute was missing
one argument which was causing the script to fail.
Signed-off-by: Zhani Baramidze <jbaramidze@meta.com>
Add `FILE` typedef in the `stdio.h` so that when doing
`FILE *file` definition checkpatch doesn't complain about
the position of the '*' and fail in CI.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
In order for the --recover option to work properly on the nRF54H20, it
requires executing it for both cores, the radio and the application one.
Extend the recover_target() function so that it does so for both 53 and
54H20.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Noticing many PRs that wait too long in the queue although once of the
maintainers approved with asignees set to other maintainers.
This changes the current behavior of picking the first maintainer in the
list and assigning to them only, instead we assign to all maintainers of
the main area being changed.
Who ends up driving the PR to a mergeable state is then to the
maintainers and they can unassign/assign based on availability.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
These clock selection Kconfigs should have been deprecated for
more than 2 releases, remove them:
- `CONFIG_COUNTER_RTC_STM32_CLOCK_SRC`
- `CONFIG_COUNTER_RTC_STM32_CLOCK_LSI`
- `CONFIG_COUNTER_RTC_STM32_CLOCK_LSE`
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Since b53a792ff0 Zephyr has the ability to define tristate Kconfig
options. When a tristate option FOO is selected as a "module", this
results in autoconf.h defining CONFIG_FOO_MODULE, not CONFIG_FOO.
This patch allows the check_compliance script to also accept references
to a Kconfig symbol ending in _MODULE if the prefix is defined.
Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
Execute 'zephyr.exe' in application build directory as 'current working
directory' (cwd). This makes sure that native_sim specific drivers (like
flash simulator with file backend in 'flash.bin') are using unique context
for external resources with relative paths.
This fixes executing native_sim tests in twister with flash simulator.
Previously a shared 'flash.bin' was used for all executed 'zephyr.exe'
processes in twister. After this patch a unique 'flash.bin' file is used
for each tested sample, since those 'flash.bin' is placed in application
build directory instead of twister root directory.
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
`west build` can be invoked without specifying the source directory when
being invoked from the source directory itself.
When using `west build` for incremental builds, then the build command
will examine the CMake cache to determine the application dir by using
the value of CMAKE_HOME_DIRECTORY.
With sysbuild, this leads to the wrong assumption that the sysbuild
itself is the application to build.
Instead, have west build look for APP_DIR which points to the correct
source dir when sysbuild is used. Use APPLICATION_SOURCE_DIR when
APP_DIR is not set, as this indicates a no-sysbuild build.
Keep CMAKE_HOME_DIRECTORY behavior as last fallback mechanism.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Stop Twister if there are too many backup copies of the output
directory already.
Before this fix, Twister silently kept artifacts from the last run,
unless `--clobber-output` was explicitly given.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Extend the coverage tool to handle applications that generate multiple
gcov dumps in a single execution. This can happen when the application
calls `sys_reboot`.
Handling multiple dumps enables coverage testing of exception handlers.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Since `writer.py` is the one writting the SPDX file, it should normalize
the name field and not `walker.py` which generates the SBOM components.
Signed-off-by: Thomas Gagneret <thomas.gagneret@hexploy.com>
Improve the SPDX with the current values:
- URL: extracted from `git remote`. If more than one remote, URL is not
set.
- Version: extracted from `git rev-parse` (commit id).
- PURL and CPE for Zephyr: generated from URL and version.
For zephyr, the tag is extracted, if present, and replace the commit id for
the version field.
Since official modules does not have tags, tags are not yet extracted for
modules.
To track vulnerabilities from modules dependencies, a new SBOM,
`modules-deps.spdx` was created. It contains the `external-references`
provided by the modules. It allows to easily track vulnerabilities from
these external dependencies.
Signed-off-by: Thomas Gagneret <thomas.gagneret@hexploy.com>
With many tests having 10s or scenarios and variants, anytime we make a
change to a test right now, we end up building all scenarios on all
platforms which ends up in multiple 10s of the thousands of instances
that need to run on 30 or 40 runners blocking CI for hours. We do not
really need that, a test needs to be smart about its coverage and not
rely on boiling the ocean to catch, mostly build errors that are
platform specific.
Change this to do the normal coverage we get on push events.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
