Reimplement atomic code in inline assembly. This can improve
optimisation, and avoids potential architectural problems with using
LDREX/STREX intrinsics.
API further extended:
* Bitwise operations (fetch_and/fetch_or/fetch_xor)
* fetch_add and fetch_sub (like incr/decr, but returning old value -
aligning with C++11)
* compare_exchange_weak
* Explicit memory order specification
* Basic freestanding template overloads for C++
This gives our existing C implementation essentially all the functionality
needed by C++11.
An actual Atomic<T> template based upon these C functions could follow.
On NUC472, on wake-up from power-down mode, we may meet hard fault or some other
unknown error. Before its cause is found, we enter idle mode instead for a workaround.
To simulate power-down mode with idle mode, we also disable us_ticker during
power-down period.
Instead of user defined symbols in assembly files or C files,
use linker scripts to add heap and stack - this is inconsistent
with ARM std linker scripts
Nuvoton M2351 was including generic core_armv8mbl.h from CMSIS - we
need it to be more specific to identify the specific core for
wait_ns. Change to core_cm23.h.
Port of NUC472/M487 crypto AC mgmt work to M2351:
1. Choose mutex to synchronize access to crypto non-SHA AC
2. Choose atomic flag to synchronize access to crypto SHA AC
1. For SHA AC, use atomic flag to manage its ownership.
(1) Nuvoton SHA AC doesn't support SHA context save & restore, so S/W
SHA fallback has been supported before. To make non-blocking 'acquire'
semantics clearer, introduce 'try_acquire' to substitute for 'acquire'.
(2) No biting CPU due to mechanism above.
(3) No deadlock due to mechanism above.
2. For AES/DES/ECC AC, change to mutex to manage their ownership.
(1) Change crypto-misc.c to crypto-misc.cpp to utilize C++ SingletonPtr
which guarantees thread-safe mutex construct-on-first-use.
(2) With change to crypto-misc.cpp, add 'extern "C"' modifier to CRYPTO_IRQHandler()
to avoid name mangling in C++.
(3) No priority inversion because mutex has osMutexPrioInherit attribute
bit set.
(4) No deadlock because these AC are all locked for a short sequence
of operations rather than the whole lifetime of mbedtls context.
(5) For double mbedtls_internal_ecp_init() issue, it has been fixed in upper
mbedtls layer. So no need to change ecc init/free flow.
With support for checking H/W UART initialized or not, we can simplify stdio management:
1. When serial_init(&stdio_uart) calls in, just set the 'stdio_uart_inited' flag.
2. When serial_free(&stdio_uart) calls in, just clear the 'stdio_uart_inited' flag.
Except above, we needn't make special handling with 'stdio_uart'.
The same H/W UART may be shared by multiple serial_t objects. This fix tries to avoid
re-configuring the same H/W UART in serial_init() when there are multiple serial_t
objects constructed. To re-configure UART, call serial_baud() and serial_format()
explicitly. This can avoid confusion when e.g. a newly constructed serial_t object
changes baudrate unexpectedly in serial_init().
RAM/ROM sizes in tools were updated to report entire device size, and in M2351
they were used earlier to report secure/non-secure partition size.
M2351 files are updated to take full RAM/ROM device size and derive secure
and non-secure partition size based on that.
`flash_get_erase_value` is added in hal/TARGET_FLASH_CMSIS_ALGO/flash_common_algo.c
duplicate copy of same function in targets/TARGET_NUVOTON/TARGET_M2351/flash_api.c
resulted in build issues with secure build
--legacyalign, --no_legacyalign are deprecated from ARMC6 compiler, in order to
remove deprecated flags all linker files (GCC and IAR as well to have uniformity)
should strictly align to 8-byte boundary
Set the ISR stack to be 1KB. https://github.com/ARMmbed/mbed-os/pull/7238
Set the heap size to 3KB(2KB + overhead + spare) so that atleast 2KB free ram is
available for testing.
With dynamic heap size, explicit size is not required. IAR 7.8 supports
static heap, hence the change is needed in IAR linker files.
Replace wait_us with nu_busy_wait_us in lp_ticker since wait_us is not allowed in sleep test
which would suspend us ticker layer on which wait_us relies. nu_busy_wait_us is implemented
by calling us ticker HAL API directly rather than relying on us ticker layer.
If us_ticker/lp_ticker is scheduled and then the interrupt is disabled, the originally scheduled
interrupt may still become pending. If this occurs, then an interrupt will fire twice on the next
call to us_ticker_set_interrupt/lp_ticker_set_interrupt - once immediately and then a second time
at the appropriate time.
This patch prevents the first interrupt by clearing interrupts in
us_ticker_set_interrupt/lp_ticker_set_interrupt before calling NVIC_EnableIRQ.
1. Modify PinMode enum to fully support GPIO I/O modes.
2. Translate input pull mode/direction to I/O mode, where H/W doesn't support
separate configuration for input pull mode/direction.
3. Allow for configuring I/O mode in addition to input pull mode.
Martin Kojtal
Kevin Bracey
ccli8
Martin Kojtal
Shrikant Tudavekar
Cruz Monrreal
deepikabhavnani
Przemyslaw Stekiel
Russ Butler
cyliangtw
Senthil Ramakrishnan
David Saada
Anna Bridge
Amanda Butler
Yossi Levy
Oren Cohen