Typically when adding a unit test directory to a CMake project a check
will be used to ensure the subdirectory is added only if the following
are true:
* The BUILD_TESTING option is set to ON.
* The current CMake project is the top-level project.
The reason being, if a downstream project includes our project they
generally don't want to build our unit tests.
In mbed-os, we do correctly specify the above condition before adding
the central UNITTEST subdirectory, which fetches googletest and adds the
"stub" libraries the unit tests depend on. However, we only check if
CMAKE_CROSSCOMPILING is OFF (or undefined) before actually adding the
unit tests. This mismatched logic would lead to unexpected build
failures in various scenarios. One likely case could be: a downstream
project including mbed-os happens to set CMAKE_CROSSCOMPILING to
OFF/undefined for any reason (possibly to build its own unit tests).
mbed-os would go ahead and attempt to build its tests without fetching
googletest or adding the required stub targets.
To fix the issue replace the check for CMAKE_CROSSCOMPILING in the unit
tests with the same BUILD_TESTING idiom we use for adding the central
UNITTESTS subdirectory.
We get a linker warning with the recently added timing module
implementation for Mbed. This is because there is Mbed TLS also ships a
file called timing.c, which we are including in Mbed OS also. With CLI
1, we get an error about unique object files because of the similarly
named implementation files.
Object file timing.o is not unique! It could be made from: mbed-os/connectivity/mbedtls/source/timing.c mbed-os/connectivity/mbedtls/platform/src/timing.cpp
Rename the Mbed timing module implementation to timing_mbed.cpp to avoid
this naming conflict.
Fixes: b8781e527b ("mbedtls: Add an alt implementation of timing")
Fixes#14759
Fix 'netsocket: several dynamic allocation results not checked' (#14210)
add_event_listener in NetworkInterface now returns an error if the method fails. Previous attempts to add the event listener would attempt to use an unchecked standard dynamically allocated ns_list_* item.
In other cases, the dynamically allocated items will now be checked, and if unsuccessful, will return after cleaning up any outstanding issues.
TCPSocket::accept will now check that its own internally allocated new TCPSocket call will succeed, and if not, will clean up the stack resources. This should help when memory is low but an incoming connection requests a connection when the TCPSocket is listening.
When calculating ufsi, the function was relying
on the slot processed by the unicast fhss timer
callback, which can be delayed. When it happens
the slot value is wrong, and the ufsi is incorrect.
The ufsi is then used by the peer to determined
the reply channel, so the devices are thus
unsynchronized until the next uplink packet.
The compiler gave redefinition warnings if
MBEDTLS_ENTROPY_HARDWARE_ALT is defined in the the compile
definitions. A check is added to prevent this warning.
Remove two expressions from at_cellularcontexttest that do nothing. This
fixes the following two warnings.
connectivity/cellular/tests/UNITTESTS/framework/AT/at_cellularcontext/at_cellularcontexttest.cpp:61:67: warning: expression result unused [-Wunused-value] ATHandler_stub::int_valid_count_table[kRead_int_table_size]; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~^
connectivity/cellular/tests/UNITTESTS/framework/AT/at_cellularcontext/at_cellularcontexttest.cpp:64:66: warning: expression result unused [-Wunused-value] ATHandler_stub::read_string_table[kRead_string_table_size];
Many test stub functions are meant to return a value, but weren't. Clang
would generate a warning for each instance where we weren't returning
anything in a function that was meant to return a value.
warning: non-void function does not return a value [-Wreturn-type]
For a specific example, my_radio::time_on_air() is supposed to return a
uint32_t, but wasn't returning anything. We'll return a zero instead of
relying on undefined behavior.
Without this, clang 11.0.1 was generating a virtual function
implementation with a `ud2` instruction to abort at run-time, causing
some execution of some unit tests to abort.
Running main() from gmock_main.cc
[==========] Running 10 tests from 1 test suite.
[----------] Global test environment set-up.
[----------] 10 tests from Test_LoRaPHYUS915
[ RUN ] Test_LoRaPHYUS915.constructor
[ OK ] Test_LoRaPHYUS915.constructor (0 ms)
[ RUN ] Test_LoRaPHYUS915.restore_default_channels
[ OK ] Test_LoRaPHYUS915.restore_default_channels (0 ms)
[ RUN ] Test_LoRaPHYUS915.rx_config
[ OK ] Test_LoRaPHYUS915.rx_config (0 ms)
[ RUN ] Test_LoRaPHYUS915.tx_config
Process 35669 stopped
* thread #1, name = 'lorawan-loraphy-', stop reason = signal SIGILL: privileged instruction
frame #0: 0x0000000000276f73 lorawan-loraphy-us915-unittest`my_radio::time_on_air(this=0x0000000800c2b048, modem=MODEM_LORA, pkt_len='\0') at Test_LoRaPHYUS915.cpp:90:5
87 };
88
89 virtual uint32_t time_on_air(radio_modems_t modem, uint8_t pkt_len)
-> 90 {
91 };
92
93 virtual bool perform_carrier_sense(radio_modems_t modem,
(lldb) disassemble --pc
lorawan-loraphy-us915-unittest`my_radio::time_on_air:
-> 0x276f73 <+67>: ud2
0x276f75: int3
0x276f76: int3
0x276f77: int3
(lldb)
- Remove redundant cmake script as already added the CMake configuration file
- Remove redundant empty_baseline as it is no longer needed with the help of CMake configuration file
This commit introduces changes to fix a missing method, `as_entry`, and traces that were not updated properly to work with this new method.
Prior to this fix, it was not possible to use KVStoreSecurityDb as the build would fail since the `as_entry` method was not declared in the class header file.
This bug was introduced with commit 957486e0eb in PR ARMmbed/mbed-os#14198
mbedtls_stub uses memset but doesn't add the <string.h> include. This
causes build failures with clang.
Add the include so we can build unit tests using clang.
- Add a new MbedOS project in mbed os root CMake which can be used along with
BUILD_TESTING conditional check for enabling the unittest build
- Update UNITTEST CMake for setting the CMake configuration like c, cxx flags etc.,
- Add if CMAKE_CROSSCOMPILING conditional check wherever target configuration check
and toolchain configuration to avoid such configuration gets included for unittest build.
- Avoid using cmsis headers as it brings more redundant dependency into unittest build and for unittest build, all the required OS flags can provide from mbed rtos types header so replaced cmsis header with mbed os types header
Don't attempt to use default_random_seed unless both PSA and Experimental API
support are enabled. This prevents errors when the Mbed TLS NV Seed feature is
being used on platforms that have PSA enabled by default, but not Experimental
API support by default.
[DEBUG] Output: compilation terminated.
[ERROR] In file included from ./mbed-os/connectivity/mbedtls/include/mbedtls/config.h:31,
from ./mbed-os/connectivity/mbedtls/include/mbedtls/entropy.h:28,
./mbed-os/connectivity/mbedtls/platform/inc/platform_mbed.h:26:10: fatal error: default_random_seed.h: No such file or directory
26 | #include "default_random_seed.h"
Due to a known issue in Mbed TLS's architecture determination
(ARMmbed/mbedtls#1077), we get the error
error: inline assembly requires more registers than available
when compiling `bignum.c` for Cortex-M0/0+/1/M23 which do not have
the macro `__thumb2__` set by the compiler.
The workaround is to define the macro `MULADDC_CANNOT_USE_R7` which
is already defined by Mbed CLI 1 but missing in our CMake support.
FixesARMmbed/mbed-os-example-lorawan#220
mbed-nanostack should depend on mbed-802.15.4-rf to avoid a linker error
with an undefined reference to `NanostackRfPhy::get_default_instance()`.
The error occurs when `device_has: 802_15_4_PHY` is defined and the
consumer depends on mbed-nanostack in their CMakeLists.txt. Previously
we linked mbed-nanostack to mbed-802.15.4-rf, so mbed-802.15.4-rf's
usage requirements weren't forwarded to consumers who depended on
mbed-nanostack.
With the previous configuration, the consumer would have to depend on
mbed-802.15.4-rf directly to avoid an issue. This seems like a layering
violation: it appears that mbed-nanostack is "the API" and
mbed-802.15.4-rf is one of several possible implementations which are
selected based on configuration macros.
This commit changes the flow of dependencies so that mbed-nanostack ends
up with the correct symbol definitions.
Previously WisunBorderRouter start() used NetworkInterface type for mesh interface, although only
WisunInterface type is possible for the call. Added a new overloads of the start with the
WisunInterface as mesh interface type and deprecated the old ones. This makes the calls stricter
about the interface type and safer. It also allows to remove the reinterpret_cast that causes
compiler warning on ARM compiler.
Duplicate status callbacks don't bring any value and can cause
memory leaks if caller calls for this function multiple times with
same status_cb. Also remove all the instances of the status_cb when
removing event listener.
Martin Kojtal
AlbanJeantheau-silabs
Alban Jeantheau
Robert Walton
Jaeden Amero
Paul Szczepanek
Tymoteusz Bloch
jeromecoutant
Chris Swinchatt
Bora Özgen
Dustin Crossman
Donatien Garnier
Abbas Bracken Ziad
Rajkumar Kanagaraj
Anna Bridge
Yahya Tawil
George Beckstein
Antti Yli-Tokola
Debdeep Saha
pennam
Martino Facchin
Jarkko Paso
Vincent Coubard
plan-do-break-fix
Peter B
Arto Kinnunen
Debdeep Saha
Lingkai Dong
Mika Leppänen
jarlamsa