Remove code block from license_check that evalutes whether a files
licenses are permissive and that there is a SPDX notice. Scancode
sometimes incorrectly attributes a single license notice to permissive
and non-permissive licenses. Removed code block results in many false
positives because it labels any file that has a "non-permissive" license
as such even when there is a permissive license.
Add function spdx_check to scancode-evaluate.py to improve analysis of
copyright and license notice tests.
Add unit test and dependent stubs for testing of scancode-evaluate.py.
license_check takes a JSON as argument and checks for source missing
copyright or license notices.
Each JSON represents a separate test case, and stubs are added for test case
3 and 4 where license_check looks for an spdx notice in the source file.
- GPIO xml parsing correction (#13711)
- Octo SPI support
- bug correction
- warning style correction
- new TIM_MST choice algo
- full PinNames.h file creation
- output directory is now
targets_custom/TARGET_STM/TARGET_STM32XX/TARGET_STM32XXXX
Passed GreenTea CI.
Added RZ_A2XX and GR_MANGO to targets.json.
Added RZ_A2XX to mbedlib.json in lwipstack and netsocket/ppp.
Added TARGET_RZ_A2XX directory to targets/TARGET_RENESAS.
Added TARGET_RENESAS_EMAC directory to features/netsocket/emac-drivers.
Moved TARGET_RZ_A1_EMAC directory into the TARGET_RENESAS_EMAC directory below and renamed to TARGET_RZ_A1XX.
Added TARGET_RZ_A2XX directory to features/netsocket/emac-drivers/TARGET_RENESAS_EMAC.
This applies only to `ARM_MUSCA_B1` target
When MCUBOOT repo: UPSTREAM was set as default as part of TF-M 1.1
release, few things were changed:
MCUBOOT_IMAGE_NUMBER: '2'
MCUBOOT_UPGRADE_STRATEGY: 'OVERWRITE_ONLY'
MCUBOOT_HW_KEY: 'On'(default)
Therefore the signing strategy for Non-secure image (Mbed OS)
needs to be done with its own private key, which is validated during
boot.
Signed-off-by: Vikas Katariya <vikas.katariya@arm.com>
Add a script to parse `targets.json` to identify PSA targets and ensure
mandatory parameters are configured correctly for all PSA targets.
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
The API is_PSA_non_secure_target() uses obsolete labels to detect if a
target is PSA non-secure target and is not needed anymore.
Mbed OS depends on TF-M for PSA SPM and services. TF-M is built using
it's own build system. Therefore, we don't need to differentiate secure
and non-secure targets anymore in Mbed OS as all PSA targets in Mbed OS
are non-secure targets.
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
Current logic `is_TFM_target` relies on the availability of attribute
`tfm_target_name` to identify PSA targets. The API `is_TFM_target` is
used in pytest to validate PSA target configuration which again checks
the availability of `tfm_target_name`. If a target doesn't contain the
attribute `tfm_target_name` then this check will fail instead of
catching it. Therefore, we now check for `TFM` config option in `labels`
attribute.
The API `is_TFM_target()` returns true for Mbed OS PSA targets which are
supported by TF-M also.
Add a new API `is_PSA_target()` which returns true for all Mbed OS PSA
targets.
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
Add TF-M to Mbed OS, replacing the previous PSA implementation for
TF-M-capable targets. This commit adds files imported from TF-M, without
modification. The version of TF-M imported can be found in
`features/FEATURE_PSA/TARGET_TFM/VERSION.txt`.
These changes switch to TF-M as the sole PSA implementation for v8-M and
dual core targets, with TF-M running on the secure side and Mbed OS
running on the non-secure side. Single core v7-M targets will continue
to have PSA implemented via PSA emulation, implemented by Mbed OS.
Move or remove many PSA-implementing files, as PSA will be provided by
TF-M on non-single-v7-M targets. Delete any files that are not relevant
for PSA emulation mode.
- Remove imported TF-M SPM
- Remove Mbed SPM and tests
- Remove Mbed-implemented PSA services and tests
- Remove PSA_SRV_IMPL, PSA_SRV_IPC, PSA_SRV_EMUL and NSPE.
- Replace PSA_SRV_EMUL and PSA_SRV_IMPL with MBED_PSA_SRV
- Remove any files autogenerated by
"tools/psa/generate_partition_code.py", which no longer exists.
Add new feature `PSA` to support PSA in Mbed OS.
Move the Mbed OS implementation of PSA services for v7-M targets (which
employ PSA emulation, and don't yet use TF-M) to
features/FEATURE_PSA/TARGET_MBED_PSA_SRV. Update the `requires`
attribute in TESTS/configs/baremetal.json to avoid breaking baremetal
testing builds.
Update .astyleignore to match new directory structure
Update Mbed TLS importer to place files into FEATURE_PSA
Create the following generic PSA targets:
* `PSA_Target` (Root level PSA generic target)
* `PSA_V7_M` (Single v7-M PSA generic target)
* `PSA_DUAL_CORE` (Dual-core PSA generic target)
* `PSA_V8_M` (v8-M PSA generic target)
Flatten MUSCA_NS and private MUSCA targets into public MUSCA targets.
Move mcuboot.bin to flat location (removing prebuilt folder)
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
Signed-off-by: Jaeden Amero <jaeden.amero@arm.com>
Currently, the final binary (TF-M + Mbed OS) is signed after
concatenating TF-M and Mbed OS binaries. But TF-M signs the images
separately and then concatenates them. Update the Musca B1 signing
strategy to match TF-M.
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
When python3 is enforced to build the ARM_MUSCA_A1 or ARM_MUSCA_B1
targets, it is unable to find binary utility tool scripts which are
imported from TF-M.
The reason to use the python3 environment is as follows: Mbed OS + TFM
contained a faulty boot record TLV, which failed the attestation test
(TF-M regression). The data in the boot record TLV will be included in
the generated attestation token as 1 item in the SW_COMPONENTS claim.
This data (in the boot record TLV) is pre-encoded in CBOR format at
build time and appended to the image during the image signing process
(done by the imgtool Python3 script).
Signed-off-by: Vikas Katariya <vikas.katariya@arm.com>
Give a warning rather than error if an unrecognised feature is used.
This will help compatibility when new features are added.
Signed-off-by: Darryl Green <darryl.green@arm.com>
For backwards compatibility reasons, since the latest Mbed OS tools must
be able to build any previous online-compiler-supported version of Mbed
OS, allow targets to use `default_lib` or `c_lib`. This should enable
the tools to work with Mbed OS 5 style targets.json.
- This change will cause build-system to produce both .bin and .hex files.
- Files are small and their creation is fast, thus there are no negative impact
on build time and storage requirements.
- .bin files are required for updating the device
- .hex files are very handy for post build (external to build system) custom
scripts (e.g. image signing)
- Configuration still allows selecting either .bin or .hex files, this settings is still
relevant as it specifies what file is flashed to a device for tests or by
`--flash` argument.
For backwards compatibility reasons, as the Mbed tools must be able to
build any online-compiler-supported version of Mbed, revert back to
emitting a warning when ARMC5 is used. This will enable ARMC5 to
continue working in the online compiler for Mbed OS 5 and other previous
versions. ARMC5 remains unsupported in Mbed 6.
This reverts commit 03d57b7ba2.
We cant afford to raise an exception now. As anyone using this out there do not catch it. Rather
an empty list, as it can work after this fix - won't do anything.
This is non trivial fix as the function is being used outside of this repository.
Tools rely on it to return list of targets for 2 or 5. As we removed release_version from many targets,
this broke the logic. To keep the logic as it was, without updating all tools out there now,
lets just return full set of targets - all are supported.
In case for Mbed 2, returning all targets does not make sense, but rather raise an exception here. Not supported. This avoids suprised. If you look at build api functions there are many checks for 2 or 5 so more
clean up needed to actually get release_version out of the tools.
cherry pick otherwise fails because the sources were moved under rtos/source. I manually resolved these patches. For the future, lets use these newer sha that should work.
Make the MUSCA_B1 target TF-M compatible by doing the following:
- Add flash, region definitions, and preprocessed image macros from
TF-M (at version 6e7be077eabe "Core: Add lifecycle API")
- Update the MUSCA_B1 linker script to create a flash image
compatible with TF-M.
- Update the tfm/bin_utils/assemble.py signing script to work with
preprocessed image macros rather than flat C pre-processor defines
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
Signed-off-by: Jaeden Amero <jaeden.amero@arm.com>
Make the MUSCA_A1 target TF-M compatible by doing the following:
- Add flash, region definitions, and preprocessed image macros from
TF-M (at version 6e7be077eabe "Core: Add lifecycle API")
- Update the MUSCA_A1 linker script to create a flash image
compatible with TF-M.
- Update the tfm/bin_utils/assemble.py signing script to work with
preprocessed image macros rather than flat C pre-processor defines
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
Signed-off-by: Jaeden Amero <jaeden.amero@arm.com>
Targets that use TF-M for their PSA implementation are not compatible
with exporters at this time. Explicitly block use of exporters with TF-M
using targets, for better error messages.
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
The PSA-implementing secure binary is not built using Mbed OS build
tools anymore. Instead, the TrustedFirmware-M (TF-M) build system is
used to produce the secure binary. As such, we remove PSA related hooks
from the build system, remove PSA related scripts from tools/test
folder, and also remove the psa-autogen job from travis which was
running the now unecessary and removed generate_partition_code.py.
Remove the ability to generate new PSA binaries in the old manner, where
Mbed OS implements PSA. We don't yet remove any PSA binaries or break
the currently checked-in Mbed-implemented PSA support. PSA targets
integrated in the old manner will continue working at this point.
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
Signed-off-by: Jaeden Amero <jaeden.amero@arm.com>
Remove PSA v8-M S target binaries will be built outside of Mbed OS and
added in as binaries which NS targets consume. Mbed OS no longer
implements PSA for v8-M targets, so there is no reason for it to build
PSA S targets.
Signed-off-by: Jaeden Amero <jaeden.amero@arm.com>
Add a copyright header and a shebang to the Mbed OS TrustedFirmware-M
tooling.
Signed-off-by: Devaraj Ranganna <devaraj.ranganna@arm.com>
Signed-off-by: Jaeden Amero <jaeden.amero@arm.com>
For release and develop profiles, turn on the linker's `--inline`
optimisation. This can save a couple of hundred bytes in a typical
image.
The optimisation replaces branches to small functions with the
inlined code from those functions. And it's possible that the
out-of-line functions can be eliminated.
Setting confined to release and develop builds only, as it can lead to
invalid debug information.
Display a post-build warning indicating to the user that the standard C
library was used instead of the small C library if the former is not
supported.
We disable C++ static destructors as best as possible in the various
toolchains, trying to eliminate unwanted destructor code. We don't want
to waste RAM or ROM on the C++-standard-specified behaviour of running
all statically-constructed objects' destructors in reverse order on
exit; we're never going to perform an orderly exit.
Techniques used include:
* `SingletonPtr` - makes an object be "lazily constructed" on first use,
and also eliminates its destructor. Lazy construction adds ROM+RAM
overhead.
* `__eabi_atexit` is stubbed out, preventing RAM usage by run-time
registration of static destructors.
* GCC has `exit` wrapped to kill shutdown code
* IAR has static destructors disabled in the compiler flags
Killing static destructors in the compiler is the optimum; if we only
stub out `__eabi_atexit`, the compiler is still inserting calls to it,
and referencing the destructors in those call.
Clang 8 added the compiler option `-fno-c++-static-destructors` (and the
object attributes `[[clang::no_destroy]]` and
`[[clang::always_destroy]]` for fine control).
We can hence enable that option in ARMC6 tool profiles, matching IAR.
This option appears to exist in ARM Compiler 6.11, but generates an
apparently spurious linker error about `EthernetInterface`. It works in
ARM Compiler 6.13, so this PR needs to wait until the compiler is
updated.
* Add optimised constexpr default constructor. Default construction
was previously by a heavyweight defaulted `nsapi_addr_t` parameter.
* Remove deprecated resolving constructor.
* Take `nsapi_addr_t` inputs by constant reference rather than value.
* Inline the trivial getters and setters.
* Use `unique_ptr` to manage the text buffer.
* Make `operator bool` explicit.
* Optimise some methods.
* Update to C++11 style (default initialisers, nullptr etc)
$(file > $@.in, $(filter %.o, $^)) is not supported in GNU Make 3.81.
Create the linker response file with pipe redirect from echo command.
This is tested with Cygwin make and make 3.8.1 shipped with macOS.
(cherry picked from commit 6918e6a76b)
Revert "Fixed problem with overlong command line."
This reverts commit dd02ac09a1.
See also https://github.com/ARMmbed/mbed-os/pull/12646#issuecomment-602058273
Harrison Mutai
jeromecoutant
Lingkai Dong
Anna Bridge
Martin Kojtal
Rajkumar Kanagaraj
Gabor Toth
Qinghao Shi
Roman Okhrimenko
Norbert Dávid
Hugues Kamba
Devaraj Ranganna
Dustin Crossman
RyoheiHagimoto
Raymond Ngun
Ashok Rao
Leon Lindenfelser
Vikas Katariya
Jaeden Amero
Wilfried Chauveau
Taylor Heck
Rob Walton
pea-pod
Darryl Green
Alexander Zilberkant
Marcelo Salazar
George Beckstein
Kevin Bracey
Veijo Pesonen
Volodymyr Medvid