All architectures define paddr_t as word_t, but the actual definition of
word_t is specific to every architecture's word size. For this reason, a
dedicated format specifier for word_t has been introduced, it hides
all these detail.
Signed-off-by: Axel Heider <axelheider@gmx.de>
The functions insert_region() and create_rootserver_objects() are not
used outside of boot.c, so there is no reason to make it publicly
available.
Signed-off-by: Axel Heider <axelheider@gmx.de>
Replace the macro SMP_TERNARY(getCurrentCPUIndex(), 0) by the much
simpler macro CURRENT_CPU_INDEX() that does the same.
Signed-off-by: Axel Heider <axelheider@gmx.de>
CURRENT_CPU_INDEX() is supported to return a word_t. The C parser from
the verification toolchain requires declaring word_t constants without
casting integer values to word_t.
Signed-off-by: Axel Heider <axelheider@gmx.de>
This fixes a correctness and security issue where uncached user
mappings might see old data from before the clearMemory operation.
See also the discussion on GitHub issue #481
Signed-off-by: Gerwin Klein <gerwin.klein@proofcraft.systems>
The link will probably go back to .html when the website is finished,
but this commit will avoid annoying test failures in the meantime.
Signed-off-by: Gerwin Klein <gerwin.klein@proofcraft.systems>
This switched the action from pull_request to pull_request_target,
because we need access to secrets. GITHUB_TOKEN permissions are
downgraded to PR-like permissions, and the action needs to be triggered
manually by labelling with label 'hw-test'.
Signed-off-by: Gerwin Klein <gerwin.klein@proofcraft.systems>
Also merge create_device_untypeds() and create_kernel_untypeds() into
create_untypeds() to simplify the code.
Signed-off-by: Axel Heider <axelheider@gmx.de>
CMake treats any text output that is generated by tools during the
configuration phase as important if it isn't part of a message(STATUS)
command. Output generated by hardware_gen.py often shows up as warnings
about device tree properties that are usually uninformative. Resolving
some of the warning conditions removes these messages for most
platforms.
- Setting kernel_size in hardware.yml to 0x1000 to handle cases where
the kernel only needs the first page of a device that has a
device-tree definition larger than that.
- Remove status print about each Interrupt processed as it's usually not
useful information.
- Only process IRQs for a selected kernel device if the rule for that
device has any interrupt queries. This prevents warnings for IRQ
controllers that the script doesn't know how to process when it
doesn't need to.
Signed-off-by: Kent McLeod <kent@kry10.com>
The reason is that with separate workflow files allow different
triggers. In particular, we don't want to re-run all simulations
on all `labeled` triggers, but if we explicitly skip the simulation
job for those triggers, than previous simulation runs are not
shown any more in the GitHub check status, so failed runs will be
overlooked.
This should achieve both: no unnecessary runs, and visible status.
Signed-off-by: Gerwin Klein <gerwin.klein@proofcraft.systems>
The previous trigger would start duplicate builds for each new
label added to a PR. This commit locks this down a bit more so
that builds only run when the trigger label is added or on other
triggers when the label is present.
Signed-off-by: Gerwin Klein <gerwin.klein@proofcraft.systems>
The action is triggered on push to master or when the label "hw-build"
is present on pull requests.
Signed-off-by: Gerwin Klein <gerwin.klein@proofcraft.systems>
Make the definition for NULL generic, so it can be used in constants
that are shared by C and assembly code.
Signed-off-by: Axel Heider <axelheider@gmx.de>
- Provide the macro ULL_CONST() for 'unsigned long long' constants, and
use it where applicable.
- Add a verbose explanation why the 'unsigned long long' type is used
for time constants.
Signed-off-by: Axel Heider <axelheider@gmx.de>
- Reorder the macro definitions to ensure things are define before they
are used.
- provide a verbose explanation why the UL_CONST() macro is needed.
- make BIT() macro is defined generic by using UL_CONST().
Signed-off-by: Axel Heider <axelheider@gmx.de>
The SimplExportAndRefine session is one of the steps of binary
verification and only available for ARM and RISCV64.
Signed-off-by: Gerwin Klein <gerwin.klein@proofcraft.systems>
This action will run the C proofs when the label `proof-test` is added
to a pull request. Intended to be used when the preprocess test has
failed and the PR is ready to check proofs.
Signed-off-by: Gerwin Klein <gerwin.klein@proofcraft.systems>
The top phys addr is exclusive, not inclusive. On 64-bit platforms there
is a trivial way to fix this as long as the whole address space is never
used. For 32-bit platforms the fix will be more complicated, so this is
postponed.
Signed-off-by: Axel Heider <axel.heider@hensoldt-cyber.de>
This adds support for the ODroidC4. The ODroidC4 is just a beefier
ODroidC2 with some bits moved around. The devices we care about here are
all essentially the same except the cores are A55s instead of A53s.
Signed-off-by: Curtis Millar <curtis@curtism.me>
This adds basic support for the ARM Cortex A55 cpu core as is used in
the Amlogic S905x3 that is found in the ODroid C4.
Signed-off-by: Curtis Millar <curtis@curtism.me>
Also make the output more verbose in case of errors, which is
helpful when porting the kernel to new platforms.
Signed-off-by: Axel Heider <axelheider@gmx.de>
