Non-SOF build does not have extended manifest data for
rimage to process, which might result in rimage error.
So do not do extended manifest during signing when not
doing SOF builds.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Now that child nodes are handled by edtlib there's no need to parse
child nodes in check_init_priorities anymore.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
The current EDT graph logic only use properties directly under a
specific node to add dependencies. For nodes properties in
child-bindings, this means that the child phandles are only linked by
the child node itself, which does have an ordinal but no corresponding
"sturct device" in the code, causing those dependencies to be silently
ignored by gen_handles.py.
Fix that by adding the recursive logic to visit child bindings when
present, which causes all child node property handles to be linked to
the parent node.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
The NXP S32 Debug Probe is a JTAG-based probe that enables debugging on
NXP S32 devices. This probe is designed to work in conjunction with NXP
S32 Design Studio and this runner offers a wrapper to launch a debug
session from cli.
`flash` command is not implemented at the moment because presently there
are no zephyr boards that can make use of it and test it.
Signed-off-by: Manuel Argüelles <manuel.arguelles@nxp.com>
Allow the runners to pass extra keywords arguments to both the server
and client subprocesses calls, as can be useful for instance to set
a specific execution environment.
ZephyrBinaryRunner.check_call() already supports passing kwargs to
the inner subprocess call but this is currently not possible when
starting server/client processes. So pass through the kwargs to each of
the client/server subprocess calls.
Signed-off-by: Manuel Argüelles <manuel.arguelles@nxp.com>
Allow the runners to override the default search path, e.g. when
provided through command line. The default is to search for the program
binary on the system PATH.
This is useful when the runner allows to optionally override the search
path of the tool, in the case there are multiple versions or
installations, and not necessarily the tools path are present in the
system PATH environment variable. For example:
`tool = self.require(tool_name, path=args.tool_path_override)`
Signed-off-by: Manuel Argüelles <manuel.arguelles@nxp.com>
Do not break processing when incorrect coverage data are received.
Instead, report failures but still process valid data.
Include coverage processing status within twister retrun code.
Signed-off-by: Piotr Kosycarz <piotr.kosycarz@nordicsemi.no>
When coverage is enabled, gather console logs longer
to receive gcov data which are send after test output.
Signed-off-by: Piotr Kosycarz <piotr.kosycarz@nordicsemi.no>
Define the ZEPHYR_{MODULE_NAM}_MODULE option for all available modules.
Right now only modules defined as 'kconfig-ext' has this symbol defined
which may cause problem if the build wants to check if a module not
using this options is available.
This allow us to do something like:
comment "mbedTLS module not available."
depends on !ZEPHYR_MBEDTLS_MODULE
or make a feature depends on the module be present.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
The arg --testsuite-root was copied from twister. When it is used
for test_plan.py it will be propagated to twister calls. This allows
to make alternative test locations (e.g. from another repo) to work
with test_plan.py
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Add --alt-tag arg for test_plan.py script. User can use it and point
to an alternative file with tag-directories relations. If so, such
file will be used instead of the default one.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Expand test_plan.py args with --ignore-path. This allows to provide an
alternative lists of patterns for the script.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
The test_plan.py script has a path to repository to be scanned for
changes hard coded to zephyr. This patch separates zephyr path from
such repository's path and adds an arg to pass repo to scan
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
An option --no-detailed-test-id was added to twister to help align
names for test outside of zephyr tree. This commit add this arg
to test_plan.py script which is then propagated to twister.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Align with native_simulator's upstream main
d32b2504ad2b6d6d541bc71a0f9b16ab7b6a3831
Which includes:
* d32b250 Minor format fix
* 7b4f35b native HW counter: Provide new API to reset and control
* 4f815cb INT CNTLR: Bugfix for more than 32 interrupts
* 1d36254 Provide new 64 version of nsi_find_lsb_set
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Cleanup of script and change how we assign PR in general:
Make sure that when multiple areas are changed, including platform or
driver changes, we set assignee of the general areas and not the
platform maintainers. For example, if I change 100 SoC files and touch
the kernel file, the assignee is set to the kernel maintainer.
First, we should as much possible try and split such changes and not
have common area changes with platform changes at the same time. Second,
it is important that such changes which affect everyone are reviewed and
approved by the maintainer of the general areas rather than leave them
to the platform maintainer.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
This commit removes the needs or generating isrList.bin temporary file.
Now gen_isr_tables.py script access the required section directly in
elf file, that was accessed by the script anyway.
It simplifies the building removing one step.
Signed-off-by: Radosław Koppel <radoslaw.koppel@nordicsemi.no>
The current description of the relationship between the tests filter
and the arch/platform allow/exclude was actually incorrect.
Let's fix it.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Very verbose log message without too much context that should be a debug
message rather than a info().
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
An urforseen error was introduce with #62713. When "--device-testing"
or "--filter runnable" is used in CLI, twister will skip tests which
cannot be executed (not just built) on a given platform. If a given
platform is among integration platforms it will cause an error.
However, it shouldn't be the case. Such skip is intentional and
shouldn't be an error. Fixed by assigning an existing Filter.CMD_LINE,
which is exempted from being turned to error.
fixes: #63844
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Reset TestInstance status when ProjectBuilder.run() is called
to run the test after its successful build, so the build status
will not propagate as the run status if the run fails.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Fix the Twister Console harness ordered 'multi_line' and 'one_line'
pattern matching to treat the ztest as failed when not all of the
expected patterns were found in the console output, but the ztest
application itself reports 'PROJECT EXECUTION SUCCESSFUL'.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
When we have an empty Devicetree, ie,
```
/dts-v1/;
/ {
};
```
The node's dep_ordinal is never initialized because the node graph is
empty. This ends up with invalid ordinal tokens (-1) in
devicetree_generated.h which in turn produce some cryptic compiler
errors, see e.g.
```
error: pasting "dts_ord_" and "-" does not give a valid preprocessing
token
95 | #define Z_DEVICE_DT_DEV_ID(node_id) _CONCAT(dts_ord_,
DT_DEP_ORD(node_id))
...
include/zephyr/devicetree.h:2498:41:
note: in expansion of macro 'DT_FOREACH_OKAY_HELPER'
2498 | #define DT_FOREACH_STATUS_OKAY_NODE(fn)
DT_FOREACH_OKAY_HELPER(fn)
|
^~~~~~~~~~~~~~~~~~~~~~
include/zephyr/device.h:1022:1:
note: in expansion of macro 'DT_FOREACH_STATUS_OKAY_NODE'
1022 |
DT_FOREACH_STATUS_OKAY_NODE(Z_MAYBE_DEVICE_DECLARE_INTERNAL)
```
(devicetree_generated.h)
```
...
#define DT_N_ORD -1
#define DT_N_ORD_STR_SORTABLE 000-1
...
```
This patch makes sure root node is always inserted (without any target)
so that it gets initialized later.
Discovered as part of
https://github.com/zephyrproject-rtos/zephyr/pull/63696
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Make `kconfig.py` not crash when any one of these settings is undefined:
* DEPRECATED
* EXPERIMENTAL
* WARN_DEPRECATED
* WARN_EXPERIMENTAL
While these will continue to exist in Zephyr, they shouldn't be strictly
required for the script to work. One situation in which they could go
missing is when a user creates an application-specific Kconfig root, but
forgets to source "Kconfig.zephyr" inside it - not that this is expected
to work anyway, but it could at least raise a more intelligible error.
Note that when WARN_DEPRECATED is undefined, selecting DEPRECATED will
always produce warnings. Same with WARN_EXPERIMENTAL and EXPERIMENTAL.
This matches the fact that the WARN_* symbols normally have `default y`.
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
There might be situations where people are running the coreutils version
of stty on macOS, hence the need for being smarter at detecting when
that might be the case
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
The runnable output from the build system is the
exe file. In native_posix the elf happend to also be
runnable, but this is not the case in general,
and not for native_sim or the nrf5*bsim boards.
So lets use the exe instead.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Add the exe fle to the runnerconfiguration class,
so we can use it from runners which will need it.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
PR #63195 has managed to change code under test without
triggering those tests.
This change should remedy that by creating a new workflow,
based on twister_tests.yaml, that runs tests in
scripts/tests/build_helpers if build_helpers are modified.
Such a workflow should be easily extendable if we ever have more
non-Twister things to test in scripts/pylib.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
PR #63195 has changed how Domains work without
triggering necessary tests, thus breaking them.
This commit fixes the domains tests.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
If flash_order were to be missing from the YAML, the Domains init
would crash instead of substituting an empty list.
This commit fixes that.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Fix the correct path to net-tools directory.
Also default to Bash instead of /bin/sh as we use
Bash syntax.
Signed-off-by: Seppo Takalo <seppo.takalo@nordicsemi.no>
Currently, the following macros will only work for 0 - 255:
- `IS_EQ`
- `UTIL_X2`
- `UTIL_INC`
- `UTIL_DEC`
- `LISTIFY`
This patch increases their limit to 4095.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
By requiring a full-on YAML fragment, the Domains class can rely on
pykwalify to do the heavy lifting with validating types and structure.
Remove support for loading domains from a dictionary, so that we don't
ever have to replicate the same checks on the input, and we can instead
have a single entry point.
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
Do extra checks to reject inconsistent `domains.yaml` contents early:
default: app
build_dir: /path/to/build
domains:
- name: image_1
build_dir: /path/to/build/image_1
- name: image_2
build_dir: /path/to/build/image_2
flash_order:
- image_1
- image_2
- unknown
This example lists "image_1" and "image_2" as known domains, but an
entry for the default "app" is missing, which should never be allowed.
A valid `domains.yaml` file shall also not contain extra domains under
`flash_order`, such as "unknown", which doesn't appear under `domains`.
Either of these cases can now be caught while initializing an instance
of the Domains class.
It follows that a valid `domains.yaml` file must always contain a list
of domains with at least one entry - the default image. Since `default`
is a required key in the YAML schema, `domains` should be required too.
Thus, empty lists will be rejected by pykwalify, so the questionable
warning for "no domains defined; this probably won't work" can be axed.
Fixes#63166
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
When using a quarantine file with more than 512 unique
entries, than time of matching quarantine increases
significantly. This is because regexp cache size is
512. Add precompiled regexp entries to the quarantine
as a fix.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
All handlers except QEMU were right-stripping console lines before
checking them. In certain cases where the regex is expecting the
whitespace, this could make the test fail on qemu when it is passing
on something else. Change QEMUHandler to us .rstrip instead of .strip()
Signed-off-by: Øyvind Rønningstad <oyvind.ronningstad@nordicsemi.no>
The network core needs to be recovered first due to the fact that
recovering it erases both the network and application cores' flash
memory. This means that the little flash image that is programmed during
the recover operation would be lost for the app core if we inverted the
order.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
In PR #62947 a timeout variable was replaced with a method,
but this timeout was used in another place to start
pytest subprogram. This commit provides a fix.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
This provides miscellaneous quality of life improvements:
- We couldn't use dataclasses when this class was originally written.
We can now, so move to dataclass to avoid having to write
__repr__().
- Add missing validation for the advertised commands
- Add missing documentation for the 'file' capability
Signed-off-by: Martí Bolívar <mbolivar@amperecomputing.com>
Test suites names are not being handled uniformly for tests not
in zephyr tree. Their names depend on -T arg used in twister's CLI.
The newly added options allow to select if twister should add paths
to suite names. This is needed if test plans are to be used for tests
outside of zephyr tree.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
"Path" is a path (relative to zephyr) to a directory with test suite's
definition. Such entry helps to locate a given test.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Domain module was not yet covered by explicit unit tests.
This commit adds a new test file to cover that gap.
It achieves 100% coverage at the time of creation.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
In test-only mode with a hardware map, all tests with fixtures were
skipped, even when the hardware map has appropriate fixtures defined.
TestPlan.load_from_file had the check for fixtures missing, but actually
best is that the TestInstance checks the fixtures in the hardware map.
Signed-off-by: Manoel Brunnen <manoel.brunnen@ithinx.io>
Twister allows us to control maximum execution time for each test
with timeout value in test's .yaml configuration and
platform level timeout multiplier which allows us to tweak
timeout value for specific platform.
However, sometimes we want to additionally adjust tests timeouts
when running twister. This is especially useful in case of
simulation platform as simulation time may depend on the host
speed & load, we may select different (i.e. cycle accurate but
slower one) simulation method, etc...
Let's introduce global (twister-level) timeout multiplier option.
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Signed-off-by: Evgeniy Paltsev <PaltsevEvgeniy@gmail.com>
There is clearly an issue with current timeout value
timeout into mocked_instance as this value is used only
once before this mock happen, so this mock is actually useless
and we always use default value. Fix that.
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Signed-off-by: Evgeniy Paltsev <PaltsevEvgeniy@gmail.com>
The experimental status of the PPP L2 is long overdue so
it can be removed as the component is working fine.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
The west bindesc command currently crashes when no subcommand is
given. This is because the subcommand was not marked as required.
This commit fixes the issue.
Signed-off-by: Yonatan Schachter <yonatan.schachter@gmail.com>
Some of the runners in the tree have been adding their own,
class-specific versions of a switch to instruct the runner to reset or
not the device after flashing.
In order to better support multi-image builds that require more than one
flash operation, introduce a new --reset,--no-reset command-line
parameter that is part of the RunnerCaps so taht this functionality can
be accessed in a standardized manner.
Implementations for the new parameter are provided for the runner
classes that were already configurable in this regard.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
Added the bindesc command to west, for working with binary
descriptors. Currently it supports dump, list and search
subcommands, for bin, hex, elf and uf2 file types.
Signed-off-by: Yonatan Schachter <yonatan.schachter@gmail.com>
This adds support for using coredump with Xtensa DC233C core,
which are being used by qemu_xtensa and qemu_xtensa_mmu.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Adds a test for the data sharing retention feature of MCUboot
configuration to an application.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
Adds a bootloader configuration interface which allows for a
bootloader (e.g. MCUboot) to set configuration in a shared data
area which is then read by the application.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
The recent support for BUILD_ONLY images was implemented by excluding
them from `domains.yaml`, in order to crudely prevent them from being
picked up by `west flash`. Arguably, this is incorrect or unexpected,
because the sysbuild documentation defines a "domain" as:
Every Zephyr CMake build system managed by sysbuild.
Another consequence is that, given a build-only `<image>`, this makes it
impossible to pass `--domain <image>` to `west flash`, `west debug`, and
ironically `west build`.
To fix that, `domains.yaml` should again represent all domains, and the
build-only ones should be indicated in another way. Enter `flash_order`:
a new top-level key in the domains YAML schema. It contains the default
sequence of images used by `west flash`, where the build-only images are
excluded, and the order is influenced by `sysbuild_add_dependencies()`.
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
Added application based on SMP Server Sample. Application is built together
with MCUboot using sysbuild and is flashed onto device in one step.
Tests are automated with pytest - new harness of Twister.
The image for upgrade is prepared using west sign command
then is uploaded by mcumgr into device and tested.
Automated scenarios to test upgrade (image upload, test, revert, confirm),
to test downgrade prevention mechanism and to test upgrade with image,
that is signed with an invalid key.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
Add support for setting reviewers as well for module PRs, in addition to
the assignees.
Note that this still only works on repositories with an assignee on
record since it does not seem to be possible to query for PRs that have
no reviewers.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Currently, when a Robot test is run, the test case status value is
reported as `None`, which causes the XML to contain information about
the test being skipped due to misconfiguration.
This commit fixes the value reported in Twister XML test result by
assigning the test result to the `status` variable of a testcase
instance after the test has finished running.
Signed-off-by: Franciszek Pindel <fpindel@internships.antmicro.com>
Signed-off-by: Piotr Zierhoffer <pzierhoffer@antmicro.com>
The C standard specifies 'hh' as a length modifier indicating an integer
conversion specifier applies to a signed char or unsigned char
argument. Python doesn't do that, so replace relevant %hh with %h.
Also fix the handling of %ll so that it applies to all integer specifiers
as well.
Signed-off-by: Keith Packard <keithp@keithp.com>
Bump number of cores limit in mdb west runner.
As we are here - adjust core number in runner test closer to
the limit.
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Signed-off-by: Evgeniy Paltsev <PaltsevEvgeniy@gmail.com>
This is coming from devicetree and corrosponds to what we have in the
dts/bindings/vendor-prefixes.txt file.
This will allow for static filtering, especially with twister, i.e. no
need to build anything to know the vendor of a board
All of the vendor data was extracted automatically from the devicetree,
so some platforms might not have the right vendor or no vendor at all
right now, we need to fix some of the DTS information or do this
manually to get this 100% correct. But we are close.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Add a new option --vendor which allows filtering by vendors tracked in
the board/platform yaml file. The vendor string is compatible with DTS
and is what we have in dts/bindings/vendor-prefixes.txt.
Providing multiple vendors is also supported.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
We don't need to have two different sets of Zephyr-specific default
values overriding one another, it's confusing.
Note this commit makes NO functional change, the effective defaults stay
the same. It does however make it easier to change defaults in the
future.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
Rework check_init_priorities to use the main executable file instead of
the individual object files for discovering the devices.
This should make the script more robust in case of stale files in the
build directory, and also makes it work with LTO object files.
Additionally, keep track of the detected init calls, and add a handy
"-i" option to produce a human readable print of the initcalls in the
call sequence, which can be useful for debugging initialization problems
due to odd SYS_INIT and DEVICE interactions.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Only impacts on twister-pytest scenarios. Exec command is waiting for
the prompt message, but in some cases, e.g. after reseting the DUT, prompt
can be followed by another strings. Solution is to clear internal serial
buffer and to not request the prompt string before executing a command.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
Sometimes you might need exec_command() to return an value from
running device. If the output contains shell prompts or logger
output, it is harder to capture the value.
Filter away the lines we thing belongs to the shell prompt or
normal logging output.
Signed-off-by: Seppo Takalo <seppo.takalo@nordicsemi.no>
Skips are turned to errors on integration platforms. However, if
a user (or a CI) selects toolchain other than zephyr, unexpected
errors due to skips can appear. With this commit skips due to
toolchain incompatibilities are not treated as errors.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
If coverage is provided by a different platform not in the integration
list, it should not be reported as error.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Remove dead and erroneous (DUT is not a dict) if branch. Simply use the
self.options.platform list. Which is filled by either the hardware map
or the '--platform' option.
This change addresses issue #62723 and #62560.
Signed-off-by: Manoel Brunnen <manoel.brunnen@ithinx.io>
Before only with --integration such skips were changed to errors.
This hinders issues when twister is call with --all flag in CI
and new skips on integration platforms are not caught. Then those
skips cause errors in other PRs when --integration was used.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
The value of emulation_platforms is a bool corresponding to
--emulation-only CLI option. Therefore, it should be registered
as Filters.CMD_LINE. Without it, twister raises errors when an
integration platform is descoped due to this option.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Added an exception to scl.py module, to not process an empty
yaml file, because of error from pykwalify. New exception
is handled when processing quarantine files, to allow
an empty file.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
After the handler.py module has been split up more,
we can update the relevant tests.
100% coverage achieved.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Simple refactoring aiming to reduce the average method length.
Minor corrections of the handlers module, removing wholly unused
variables, etc.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Expansion of the tests related to handlers.py
Achieves over 85% coverage.
Implemented most of golowanow's suggestions.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Allow to specify a list of pytest directories, files or subtests
with pytest_root keyword in test yaml.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
Align with native_simulator's upstream main
64b8be6311a61df2c63987813e8058560c8dc49c
Which includes:
* 64b8be6 Make: Give option to add extra localization parameters
* 05bc3ce Generalize code for N CPUs/MCUs
* 489069b Makefile: Generalize for N embedded images
* a4c817e runner->embedded trampolines: Improve definitions
* 8d8dd29 nsi_utils: Provide debracket macro
* 6b4956e Add optional embedded test hook definition
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Add a check to ensure that every module has a corresponding maintainers
file entry, ensure modules are not added with no recorded point of
contact.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Fixtures in pytest-twister-harness plugin are moved to one
file to simplify adding new fixtures in the future - no
need to add pytest_plugins entry and register asserts.
Moved also shell fixture from sample dir, because that
fixture can be reused in new tests.
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
This commit introduces all the necessary changes for
enabling the usage of shared interrupts.
This works by using a second interrupt table: _shared_sw_isr_table
which keeps track of all of the ISR/arg pairs sharing the same
interrupt line. Whenever a second ISR/arg pair is registered
on the same interrupt line using IRQ_CONNECT(), the entry in
_sw_isr_table will be overwriten by a
(shared_isr, _shared_sw_isr_table[irq]) pair. In turn, shared_isr()
will invoke all of the ISR/arg pairs registered on the same
interrupt line.
This feature only works statically, meaning you can only make use
of shared interrupts using IRQ_CONNECT(). Attempting to dynamically
register a ISR/arg pair will overwrite the hijacked _sw_isr_table
entry.
Signed-off-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com>
Update CFB font generator so it works with Pillow version 10. They
deprecated some methods, with no direct replacements, so the generated
fonts might be slightly different.
Signed-off-by: Jonathan Rico <jonathan@rico.live>
If pylib twister harness reads bytes that can't be decoded to Unicode,
it would raise an exception at a separate "serial reader" process, that
would fail without any warning.
This patch uses "errors='replace'" parameter for Python str.decode()
method, so that this doesn't happen.
Signed-off-by: Ederson de Souza <ederson.desouza@intel.com>
A commit breaking tests slipped past CI checks.
This commit fixes broken tests in test_testsuite.py.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Previously, only the main src/ directories was scanned for
ZTest testsuite names and tests. This allows us to place
test sources in subdirectories as well.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Add support handling quotes for configuration options in extra args by
escaping them properly instead of removing the quotes altogether. For
other options in extra_args quotes are removes as usual.
Add similar support in west build command also.
Add a unit test to check this functionality.
Signed-off-by: Chaitanya Tata <Chaitanya.Tata@nordicsemi.no>
Log the device priority on the file debug output print, this is
convenient to list out all the defined devices and their priority, even
if they don't depend on each other.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
During testing it was found that if the partition type was displayed as
msdos it could not be detected correctly, so this partition type
determination was added.
Signed-off-by: Yuchao Guo <yuchao.guo@hoorii.io>
LinkServer is NXP's tool to flash and debug on MCU's.
This patch expands support for LinkServer to MIMXRT1170.
There is one limitation with flashing and debugging the SoC's seciond
core (COrtex M4) that will be addressed with a future submission.
Signed-off-by: Yves Vandervennet <yves.vandervennet@nxp.com>
This should increase the coverage of the testsuite.py module
up to 100%.
Some tests have been commented out, as the code tested there is
incorrect. Thus the real increase is 99%.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
It is only available if CONFIG_DEVICE_DEPS=y, so the ZephyrElf class
would fail to load if it does not treat _DEVICE_STRUCT_HANDLES_OFFSET as
optional.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Before, only following way of getting test-item build was supported:
`west build -b <BOARD> -T <PATH_TO_SOURCE/TEST_NAME>`
This commit allows to also call it by:
`west build -b <BOARD> <PATH_TO_SOURCE> -T <TEST_NAME>`
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Add tests which use pytester to call pytest with pytest-twister-harness
plugin on previously defined test. This gives a possibility to test
plugin in end-to-end variant.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
According to new adapters API add new tests which verify their
correctness. Introduce some minor naming changes to describe better
their role.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Add better description of available options and add actions which
should help to prevent from passing invalid options to plugin.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Remove log.py and by this logging configuration from
pytest-harness-plugin. Instead use pytest options for setting verbosity
and format of printed logs. Thanks to this management of verbosity
level can be easier for users.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Create helper class Shell which can be used in tests which require
interaction with shell application.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Move helper.py file content to utils.py file in device directory,
where those helper functions are used.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Divide dut fixture into three smaller fixtures to make it easier
to user to change dut fixture scope or to write her/his own dut
fixture. Add initialize_log_files method to make it possible to
initialize log files per each test.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Pass pytest test name to handler.log header to make it easier to
analyze handler.log content.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Add readlines_until method to DeviceAdapter class to simplify using
adapters in tests by give user possibility to read output from device
until following conditions:
- provided regex was found
- provided number of lines was already read
- timeout was occurred
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Introduce changes into FifoHandler class to make it more safe
(especially to avoid blocking all program when FIFO queue is not opened
by other application).
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Select and place common code of three adapters (HardwareAdapter,
BinaryAdapterBase and QemuAdapter) into basic DeviceAdapter class.
Introduce new way of reading device output by run separate thread which
try to read device output and place it into internal python queue.
Thanks to this, now it is possible to create readline method for all
adapters, which can be unblock when timeout occur.
Collect all common steps which have to be done before setup device in
launch method. The same was done for teardown operations which were
placed into close method.
Additionally some protection mechanisms were introduced to prevent for
undesirable side-effects when user could try to launch to already
launched device or try to send some data to disconnected device.
iter_stdout method was replaced by two new methods: readline and
readlines. To make it possible to remove all read output from internal
buffer (queue), clear_buffer method was introduced.
Also unit tests were rewritten to work properly with current version
of adapters.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Change name of SimulatorAdapterBase class to more meaningful like
BinaryAdapterBase. This changes is also made to be more align with
Twister terminology.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
Add possibility of passing timeouts from Twister to pytest and unify
various timeouts used in adapters to point to one main timeout.
Signed-off-by: Piotr Golyzniak <piotr.golyzniak@nordicsemi.no>
PytestAssertRewriteWarning is raised when pytest-twister-harness
plugin is called from twister. Call register method to get rid
of that warning
Signed-off-by: Grzegorz Chwierut <grzegorz.chwierut@nordicsemi.no>
When parsing test names from symbol names, use a non-greedy match to
avoid parsing test names with __ in them.
The symbol looks like z_ztest_unit_test__SUITE__TEST. It is more likely
that a test name will contain __ than a suite name, so now something
z_ztest_unit_test__a_b__c_d__e_f will be parsed as suite:a_b
test:c_d__e_f, whereas it used to be parsed as suite:a_b__c_d test:e_f
Signed-off-by: Jeremy Bettis <jbettis@google.com>
If a test doesn't start with test_, twister currently emits an error:
suite_name.test_name: Unknown status 'started'
Following the suggestion by Maksim Masalski on PR#21191, change the
regex to exclude spaces in the test name, and remove the check that the
test_ prefix exists.
See also #21162
Signed-off-by: Jeremy Bettis <jbettis@google.com>
Now, if the rimage target (meaning rimage configuration and platform name)
is not defined in board.cmake the sign script returns fatal error.
Change this to a warning since there are configurations that are
not using 'west sign' or is used just to glue the headers of the
final image.
Also, update the documentation accordingly.
Signed-off-by: Iuliana Prodan <iuliana.prodan@nxp.com>
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
In addition to the white-box testing and unit tests,
we would like to have some black-box testing for the
console commands of twister.
This serves as a Proof of Concept for future expansion.
Special dummy tests are made for blackbox testing.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
Following PR #61471, ehl_crb board is now intel_ehl_crb.
Update the footprint test plan accordingly.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Fix an "if" that should have been an "elif". This is currently causing
the script to fallback into the "do all unassigned PRs of the day" path
when a PR is specified.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Add a -I option to process an issue and, if it's not assigned already,
add assignees based on the labels, matching the ones set in the
MAINTAINERS file areas.
Suggested-by: Henrik Brix Andersen <henrik@brixandersen.dk>
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Add a -m option to set_assignees to have the script find all unassigned
module PRs for areas that have an assignee and set the assignee on those
PRs.
This should help avoiding forgotten PRs on modules. It's implemented as
a search and meant to run periodically, so we don't have to add a
workflow for this on every individual module.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
The file kconfig.py was originally introduced by myself in:
https://github.com/zephyrproject-rtos/zephyr/pull/5569
The file was a modified version of a Kconfiglib one, so it makes sense
to keep the original license (ISC).
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Force default stdout encoding to utf-8 as otherwise ram_report/rom_report
may fail to render (ex. in CLion IDE built-in terminal).
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Add platform filtering when generating tests for manifest modules.
In the clang workflow, we do select to run only on one platform, so this
needs to be applied as well or we end up building unwanted tests using
the wrong toolchain.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
strtok_r is part of GNU99, but default Zephyr is built with C99, so,
the GNU headers are not pulled in, even if we add "--std=gnu99" it
won't override the C99 standard.
As a workaround for build we redefine the prototype, but this causes a
coding guideline warning, so, for now disable this warning
for strtok_r.
Signed-off-by: Sachin D Kulkarni <sachin.kulkarni@nordicsemi.no>
The generated stubs use retval for function return value.
However, this is too generic and is going to collide with
other code. So rename those to be more specific. Note
that the double underscore is intentional to minimize
chance of being shadow variables.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
While the code already includes some provisions to allow running outside
of a git checkout, attempting that right now causes a failure to generate
the XML report, as the JSON report creates a null `zephyr_version` field.
The reason for that is that the original code doesn't set `self.version`
when the subprocess returns a non-zero status (and also doesn't log.)
Instead of having to set `self.version` for all failure cases, initialize
it to the failure state, and log if it hasn't changed from that.
Signed-off-by: Diego Elio Pettenò <flameeyes@meta.com>
When using the hardware map and testing on device, print out the name of
the dut (from the hardware map) on the script and report it in the json
output as well.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
PCI devices are have some differences to regular nodes:
* node name specifies device/function e.g. "pcie@1,0"
* register address has a different meaning
* zero-sized register is allowed
This improves alignment with Linux DT for PCI devices
Signed-off-by: Grant Ramsay <gramsay@enphaseenergy.com>
The anytree package has introduced a breaking change in version 2.9.0 by
adding a 'size' property to the NodeMixin class. Since the TreeNode class
in size_report derives from NodeMixin and defines an attribute with the
same name, an AttributeError is raised when generating reports.
With this change, the attributes of the TreeNode class are prefixed with
an underscore to resolve the name collision and to prevent future name
collisions.
Fixes#60213.
Signed-off-by: Christian Marx <c.marx@vega.com>
Align with the latest upstream native simulator
4c595794588f9d7f67fcf0fe05c3db02892a00f9
including:
* 4c59579 Makefile: Add option to build native part
* 910f934 Makefile: NSI_EXTRA_INCLUDES option and lots of commentary
* d9bf489 cmd line parsin: Minor header refactoring
* 02f3555 cmd line cleanup: Run as NSI_TASK instead of calling expl.
* 2c88173 Split exit call in two
* 2b989b4 CPU IF change: nsif_cpu0_cleanup() to return int
* e696228 HW scheduler: Add API to get next event time
* ae0e9e8 native irq ctrl: Miscellaneous fixes and improvements
* 3fd84cd NSI_TASK: Add compile check of valid priority
* 7e09fb8 HW events: Change internal storage
And two minor updates to the native_sim board,
to align with this updated version:
* nsif_cpu0_cleanup(void) now must return an int
* We need to explicitly tell the native simulator build we want
the native components built
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
In Linux, checkpatch.pl relies on the vendor-prefixes.yaml file
to validate manufacturers in compatible strings.
In addition to the vendors defined in vendor-prefixes.txt,
the YAML file includes expressions for "prefixes which are not vendors":
these expressions do NOT define special manufacturers that may appear
in compatible strings, and are never involved as such in DTS files.
We can rather see them as bulk-definitions of JSON/YAML properties
suitable for the dt-schema tools.
OTHO, in Zephyr, checkpatch.pl relies on the vendor-prefixes.txt file,
which does not include these additional prefixes, but edtlib.EDT adds
them as hard-coded special values.
This is confusing, if not incorrect:
- the fact that edtlib.EDT (and thus its client code in the
zephyr/scripts directory) actually allows these vendors
in compatible strings is buried in the source code
- checkpatch.pl (with vendor-prefixes.txt) in Zephyr behaves neither like
checkpatch.pl (with vendor-prefixes.yaml) in Linux, nor like edtlib.EDT
(with _VENDOR_PREFIX_ALLOWED)
- Zephyr should not treat these "prefixes which are not vendors" as
valid manufacturers in compatible strings to begin with
Signed-off-by: Christophe Dufaza <chris@openmarl.org>
Currently, this script will blow up when given a symlink for a path (e.g.,
to a kconfig file).
This is annoying, as when wrapping the build system in a hermetic build
system like Bazel, Bazel likes to limit the input file set by creating
symlinks.
Resolve the path, if it is a file once resolved, then it's OK.
Signed-off-by: Jack Rosenthal <jrosenth@chromium.org>
Some architectures such as RISC-v support more than 255 interrupts
per aggrigator. This diff adds the ability to forgo the aggrigator
pattern and use a configurable number of bits for multilevel
interruts.
Signed-off-by: Joshua Lilly <jgl@meta.com>
DTC_OVERLAY_FILE was replaced with EXTRA_DTC_OVERLAY_FILE. Adjust example
usage in schema file.
Fixes: 1561a0705f ("snippets: support for EXTRA_DTC_OVERLAY_FILE and
EXTRA_CONF_FILE")
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
Commit 9b30667c77
("build: drop LEGACY_INCLUDE_PATH support")
removed scripts/utils/migrate_includes.py.
This was premature. This script is still useful to users of older
versions of zephyr that are attempting to migrate to current ones.
For example, users of v2.7 LTS who are migrating to v3.4 should
still be able to run the script to migrate their code bases.
Restore it. Maybe after next LTS we can remove it. It costs
basically nothing to keep it in the tree.
Signed-off-by: Martí Bolívar <mbolivar@amperecomputing.com>
If size is given then twister prnits the size information for the ELF
but returns one which if used in automation fails the command.
Return zero as the command is successful. Also, fix the unit test case.
Signed-off-by: Chaitanya Tata <Chaitanya.Tata@nordicsemi.no>
Let's not provide default toolchains for 'arc' intentionally: some
targets can't be built with GNU toolchain ("zephyr", "cross-compile",
"xtools" options) and for some targets we haven't provided MWDT
compiler / linker options in corresponding SoC file in Zephyr, so
these targets can't be built with ARC MWDT toolchain
("arcmwdt" option) by Zephyr build system.
Instead for 'arc' we rely on 'toolchain' option in board yaml
configuration.
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Signed-off-by: Evgeniy Paltsev <PaltsevEvgeniy@gmail.com>
This patch adds Kconfig options to select either GNU libgcc or LLVM
compiler-rt. The 'rtlib' flag is provided in a config file, so this
patch introduces 'clang_libgcc.cfg' and 'clang_compiler_rt.cfg' which
enable appropriate library. The file is selected by concatenating
the 'clang_' prefix with library name.
Signed-off-by: Patryk Duda <pdk@semihalf.com>
This is a one-line fix for edtlib, which lets gen_defines.py indicate
whether the `ranges` property exists within a given node.
Although address translation through ranges is typically automatic,
users can choose to manually inspect ranges using DT_FOREACH_RANGE(),
DT_NUM_RANGES(), and other DT_RANGES_* macros. These can be used to
implement manual translation at runtime, which is currently done for
PCIe controllers.
The only thing missing is being able to check if a node contains an
empty `ranges;`, which signifies a 1:1 translation to the parent bus.
Checking DT_NUM_RANGES() is insufficient, because it returns zero
whether or not `ranges;` is present.
It should be possible to use DT_NODE_HAS_PROP(), but it was not working,
because edtlib ignores properties which are undeclared in bindings and
don't have a default type. Add a missing PropertySpec for `ranges` with
"compound" type; it can't be "array" because it can be empty-valued.
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
Enable the possibility for boards to implement a custom `run` target in
its board.cmake to run any arbitrary commands. This is helpful for devs
who would like to add support for proprietary simulator to their boards
that can't be upstreamed.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Add an additional option to the `append` schema for appending to the
`DTS_EXTRA_CPPFLAGS` cmake cache variable, enabling finer control over
the content of devicetree files.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
For applications relocating big parts of the code with many sections,
builds were failing on Windows due to hitting the max command-line
length on that platform.
Fix this by using a file to store the dictionary passed to the python
script.
Fixes https://github.com/zephyrproject-rtos/zephyr/issues/60994.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
We have been dealing with missing and multiple binaries the same way and
both would result in a build error, which is not accurate. multiple
binaries in the build directory are fine, we just need to pick the right
one for parsing.
If we get no binaries, raise an exception and report failure, however,
if we have multiple binaries, filter intermediate artifacts out and
parse what remains.
qemu binaries generated after a run are also being filtered here. Those
are not build artificats and appear only after running in qemu. However
they have been causing issues on retries.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
When no/too many elf files are detected after a build a BuildError
exception is raised. However, it was not being counted as an issue
with a test. With the patch satuses of tests' with such exception
are reported as errors. Whithout it, twister finished without
reported errors and was getting green CI checks.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
The check currently only runs if the maintainers file itself is changed,
but that means that the check is going to miss every PR that moves
directory or delete files that can potentially trigger an error.
This check is cheap to run, just run it unconditionally.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Update the script to parse the new section naming. The ordering type
is converted from an integer to a tuple, which still compares correctly
due to the elementwise behaviour of tuple comparison.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Generate a zero padded variant of `_ORD` that is suitable for use in
linker scripts with the `SORT` property, so that `6` is correctly placed
before `24`, and so on.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Generate a zero padded variant of `_ORD` that is suitable for use in
linker scripts with the `SORT` property, so that `6` is correctly placed
before `24`, and so on.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
This is meant as a substitute for sys_clock_timeout_end_calc()
Current sys_clock_timeout_end_calc() usage opens up many bug
possibilities due to the actual timeout evaluation's open-coded nature.
Issue ##50611 is one example.
- Some users store the returned value in a signed variable, others in
an unsigned one, making the comparison with UINT64_MAX (corresponding
to K_FOREVER) wrong in the signed case.
- Some users compute the difference and store that in a signed variable
to compare against 0 which still doesn't work with K_FOREVER. And when
this difference is used as a timeout argument then the K_FOREVER
nature of the timeout is lost.
- Some users complexify their code by special-casing K_NO_WAIT and
K_FOREVER inline which is bad for both code readability and binary
size.
Let's introduce a better abstraction to deal with absolute timepoints
with an opaque type to be used with a well-defined API.
The word "timeout" was avoided in the naming on purpose as the timeout
namespace is quite crowded already and it is preferable to make a
distinction between relative time periods (timeouts) and absolute time
values (timepoints).
A few stacks are also adjusted as they were too tight on X86.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
The new stack zephyr,cdc-acm-uart driver has two separate init path, one
of which kicks in before the USB stack to start buffering log messages.
This generates a false positive in the build time priority checking,
adding a check to ignore that compatible entirely.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Parse common section and append to cmake args if available when using
--test-item option of west build.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Linkserver is a utility for launching and managing GDB servers for NXP
debug probes, which also provides a command-line target flash programming
capabilities. Linkserver can be used with NXP MCUXpresso for Visual Studio
Code.
For more information about LinkServer, please visit the LinkServer web
page (link [1] below).
This commit adds a runner to west, supporting debug and flash commands.
Documentation is also added.
[1] - LinkServer web page:
https://www.nxp.com/design/software/development-software/mcuxpresso-software-and-tools-/linkserver-for-microcontrollers:LINKERSERVER
Signed-off-by: Yves Vandervennet <yves.vandervennet@nxp.com>
Add the BT Telephony and Media Audio Profile broadcast samples.
These reflect devices that control telephone and media audio
of LE Audio.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Adds a guide on how to make out-of-tree MCUmgr function handlers
and groups, with an example showing a test implementation.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
When using twister to generate coverage with the coverage tool lcov,
allow using --coverage-formats to pick if you want lcov or html,
defaulting to both. Picking html will also use lcov, since that is
required for geninfo. This will allow callers to avoid the potentially
slow and disk intensive html reports if they only wanted the lcov info
file.
Signed-off-by: Jeremy Bettis <jbettis@google.com>
Fixes Issue #60522 by guarding member access against None parameters.
Shortens the init by using the if...else assignment.
Fixes two typos.
Signed-off-by: Lukasz Mrugala <lukaszx.mrugala@intel.com>
