GCC 14.3 is not happy when this variable is left uninitialized, although
we don't actually care about the contents, just the address.
Signed-off-by: Keith Packard <keithp@keithp.com>
Due to the constant alloc/free cycles during the stress test,
it is possible that the heap can become fragmented enough to
prevent successful allocation, and thus failing the test.
So we change the CONFIG_SYS_HEAP_ALLOC_LOOPS to a higher value
so the allocation code will try harder to find a space.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
There are lots of copy and paste error message which does not
really tell you which part fails. So make those messages more
explcitly on where it is emitting the error message.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The syscall stress test relies on timeslicing to switch between
threads, which utilizes the interrupt handling path to perform
context switching. It is also possible for kernel to switch
threads due to, for example, taking a mutex. This adds a bit of
code to the syscalls test to simulate this by yielding while
inside syscall handlers. This is to cover more scenarios so we
can catch issues earlier.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Add a loop to join the stress threads after calling abort.
This is to make sure all stress threads have already stopped
before moving on. Simply for test hygiene.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This moves the syscall torture test into another test group or
suite so that we run the individual syscall tests first before
the torture test. If individual test fails, it is easier to see
the error. Especially under SMP where multiple CPUs can report
the same error, cluttering the output.
Also rename the test to syscall_switch_stress to better describe
its purpose.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This modifies string_copy() to have an addition ID as function
argument so that it can copy into different buffers during
the multi-threaded stress test, instead of all threads writing
into the same buffer. This is to prevent a situation where
one thread writes incorrect data but being overwritten by
another thread with correct data, thus making the following
strcmp() passing for both threads. Logically this should never
happen since the same function is being called which should
give same result, but it is simply to test for weird
situations.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The TOOLCHAIN_DISABLE_WARNING/TOOLCHAIN_ENABLE_WARNING macros are easier
to read and compiler agnostic.
Signed-off-by: Tom Hughes <tomhughes@chromium.org>
Add some overlay files for the silabs xg29_rb4412a board to enable tests
on the board. Also add the platform to some testcase.yaml files.
Signed-off-by: Martin Hoff <martin.hoff@silabs.com>
The number of mpu regions that can be configured is less than four cases.
Therefore, only remove this case on cm33 cores, failed log show below:
"num_parts of 4 exceeds maximum allowable partitions (3)"
samples: kernel: mem_protect: userspace: remove mimxrt700_evk cm33 cores
The maximum number of partitions is less than the number of partitions
currently required, causing case to be unable to apply for or delete
excess regions.
Signed-off-by: Lucien Zhao <lucien.zhao@nxp.com>
Mostly a revert of commit b1def7145f ("arch: deprecate `_current`").
This commit was part of PR #80716 whose initial purpose was about providing
an architecture specific optimization for _current. The actual deprecation
was sneaked in later on without proper discussion.
The Zephyr core always used _current before and that was fine. It is quite
prevalent as well and the alternative is proving rather verbose.
Furthermore, as a concept, the "current thread" is not something that is
necessarily architecture specific. Therefore the primary abstraction
should not carry the arch_ prefix.
Hence this revert.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
This adds data cache manipulations, flushing and invalidation,
to the tests where buffer content are being written and
compared. These tests map different virtual pages to the same
physical pages, and write to one of the mapped virtual
addresses. Some SoCs may cache the virtual address separately
and writes to one virtual address will not be reflected to
another virtual address, this failing the comparison. So we
need to manually flush the cache after writing to the buffer,
and invalidating cache before reading. Note that not all reads
and writes need this treatment as some of them only needs to
test for access permissions, and not the memory content.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
For CPUs with incoherent cache under SMP, the tests to read/write
buffer (... majority of tests here) may not work correctly if
the test thread jumps between CPUs. So use the test infrastructure
to limit the test to 1 CPU.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Both tests set CONFIG_COMMON_LIBC_MALLOC_ARENA_SIZE=0 so extract
it to prj.conf. The tests require malloc arena size to be 0 to
run successful. By refactoring the config, it is now able to
with west build and still passing.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
`_current` is now functionally equals to `arch_curr_thread()`, remove
its usage in-tree and deprecate it instead of removing it outright,
as it has been with us since forever.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Amend the filtering so that the normal mem_map (with exec) test
is not going to run on Intel Audio DSP SoCs.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
With the rename of intel_adsp/ace30_ptl to intel_adsp/ace30/ptl,
the "sim" variant no longer inherit the base configuration. So
make a copy of the .conf file to explicitly target the sim
variant.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Obviously, everyone knows that there are 8 bits per byte, so
there isn't a lot of magic happening, per se, but it's also
helpful to clearly denote where the magic number 8 is referring
to the number of bits in a byte.
Occasionally, 8 will refer to a field size or offset in a
structure, MMR, or word. Occasionally, the number 8 will refer
to the number of bytes in a 64-bit value (which should probably
be replaced with `sizeof(uint64_t)`).
For converting bits to bytes, or vice-versa, let's use
`BITS_PER_BYTE` for clarity (or other appropriate `BITS_PER_*`
macros).
Signed-off-by: Chris Friedt <cfriedt@tenstorrent.com>
What is changed?
- Added a new mps3 board an552 for the soc corstone300.
The qualifier to build/run application with board mps3/an552 is
`mps3/corstone300/an552` for secure and
`mps3/corstone300/an552/ns` for non-secure.
- Added FVP variant to enable FVP testing with corstone300
and it uses the ARM FVP `FVP_Corstone_SSE-300_Ethos-U55`.
The qualifier to build/run application with FVP is
`mps3/corstone300/fvp` for secure and
`mps3/corstone300/fvp/ns` for non-secure.
- Note: the qualifier to build/run application with board mps3/an547
is now changed to
`mps3/corstone300/an547` for secure and
`mps3/corstone300/an547/ns` for non-secure.
How is it changed?
- Moved common code from mps3/an547 to corstone300.
- Renamed soc for an547 to corstone300 and added
a new soc corstone300/an552.
Why do we need this change?
- This enables FVP support and testing for corstone300.
- SOC/qualifier for mps3/an547 was renamed to reduce code redundancy
- A separate FVP variant was added for AN552 because, the TFM board
used for non-secure variant differs for FPGA and FVP.
TFM board `arm/mps3/corstone300/fvp` should be used when testing
AN552 with FVP and `arm/mps3/corstone300/an552` should be used when
testing with AN552 FPGA.
Signed-off-by: Sudan Landge <sudan.landge@arm.com>
The test_k_mem_map_unmap test requires some free physical pages
to work correctly. On qemu_x86_tiny, the physical memory is
artificially limited to test demand paging, which is 320KB as
of writing of this commit message. We also reserve 128KB of
physical memory as swapping area. And we do pin quite lot of
text and data (relatively speaking) in memory. There is not
much memory left for the test. So lower the amount of reserved
memory for paging to leave some pages for the test.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This needs special treatment because the TEST_MEM_MAP section
is placed at the end. Since there is code in TEST_MEM_MAP,
rimage thinks the whole text section spans from .text to
end of TEST_MEM_MAP, which overlaps .data and others, so
it complains. Skip the execution test to avoid this issue.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The board has been renamed from intel_adsp/ace30_ptl to
intel_adsp/ace30/ptl. So the corresponding board only configs
for tests also need to renamed.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Renamed soc from ace30_ptl to ace30.
We were previously using the wrong soc name.
The correct name is ace30.
There is only one ptl platform, but there can be several ace30 platforms.
Signed-off-by: Grzegorz Bernat <grzegorzx.bernat@intel.com>
Up until now, the `__thread` keyword has been used for declaring
variables as Thread local storage. However, `__thread` is a GNU
specific keyword which thus limits compatibility with other
toolchains (for instance IAR).
This PR intoduces a new macro `Z_THREAD_LOCAL` which expands to the
corresponding C11, C23 or C++11 standard keyword based on the standard
that is specified during compilation, else it uses the old `__thread`
keyword.
Signed-off-by: Daniel Flodin <daniel.flodin@iar.com>
For code clarity, remove unnecessary `return` statements
in functions with a void return type they don't affect control flow.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
This provides memory mappings with the ability to be initialized in their
paged-out state and be paged in on demand. This is especially nice for
anonymous memory mappings as they no longer have to allocate all memory
at mem_map time. This also allows for file mappings to be implemented by
simply providing backing store location tokens.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
When content validation fails it is rather difficult to figure out the
cause if some meaningless byte mismatches another meaningless byte.
Let's store each memory location with its own virtual address so if a
mismatch occurs when reading it back it is obvious where it comes from.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Add the ability to touch memory in the opposite direction from one loop
to the next so to exercize the LRU eviction algorithm.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Make sure k_mem_unmap() completes successfully on memory that is half
paged-out and such memory is no longer accessible afterwards.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
This is part of a series of moving memory management related
stuff out of the Z_ namespace and into its own namespace.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The board defconfig forces CONFIG_USERSPACE to be disabled
due to small number of available MPU regions which are not
enough for userspace testing.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Also any demand paging and page frame related bits are
renamed.
This is part of a series to move memory management related
stuff out of the Z_ namespace into its own namespace.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This is part of a series to move memory management related
stuff from Z_ namespace into its own namespace.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This is part of a series to move memory management functions
away from the z_ namespace and into its own namespace. Also
make documentation available via doxygen.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This renames z_phys_map() and z_phys_unmap() to
k_mem_map_phys_bare() and k_mem_unmap_phys_bare()
respectively. This is part of the series to move memory
management functions away from the z_ namespace.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Rename all Silicon Labs radio boards to always use the official kit name of
the kit containing only the radio board as board target name. Disentangle
boards using Series 1 and Series 2 SoCs that have almost nothing in common
in terms of devicetree content, but still occupied the same directory.
As part of the initial HWMv2 port, all radio boards were merged into a
single target efr32_radio, using the SoC as a board qualifier to target
specific boards. This was unfortunate, as the SoC name is not sufficient
to identify a specific board -- the same SoC may be used on multiple
different boards. This commit reinstates every radio board as an individual
board target.
Signed-off-by: Aksel Skauge Mellbye <aksel.mellbye@silabs.com>
Adding bt_tester support.
Adding bt_tester app build
with Twister for frdm_rw612 board
Fix failing tests/samples due to unsufficent MPU ressources
by releasing unnecessary MPU regions.
Twister builds added on peripheral/central ht and
enabled Power manager on those examples to cover
the CI builds with PM enabled.
Signed-off-by: Yassine El Aissaoui <yassine.elaissaoui@nxp.com>
