xPortStartScheduler calls vPortSetupTimer -> _frxt_tick_timer_init,
which enables tick timer interrupt and sets up the first timeout.
From that point on, the interrupt can fire. If the interrupt happens
while _frxt_dispatch is running, the scheduler will enter an infinite
loop. This is because _frxt_dispatch isn't supposed to be preemptable,
and the tick interrupt will overwrite some of the registers used by
_frxt_dispatch.
Note that this situation doesn't practically occur on the real
hardware, where the execution of vPortSetupTimer and _frxt_dispatch
happens quickly enough. However it can be reproduced on an emulator
if the tick period is set to 1ms.
Add an explicit call to portDISABLE_INTERRUPTS in xPortStartScheduler
to guarantee that _frxt_dispatch doesn't run with interrupts enabled.
This is similar to the esprv_intc_int_set_threshold(1); call in
RISC-V version of port.c.
Add TRY_ENTRY_CRITICAL() API to all for timeouts when entering critical sections.
The following port API were added:
- portTRY_ENTER_CRITICAL()
- portTRY_ENTER_CRITICAL_ISR()
- portTRY_ENTER_CRITICAL_SAFE()
Deprecated legacy spinlock API in favor of spinlock.h. The following API were deprecated:
- vPortCPUInitializeMutex()
- vPortCPUAcquireMutex()
- vPortCPUAcquireMutexTimeout()
- vPortCPUReleaseMutex()
Other Changes:
- Added portMUX_INITIALIZE() to replace vPortCPUInitializeMutex()
- The assembly of the critical section functions ends up being about 50 instructions longer,
thus the spinlock test pass threshold had to be increased to account for the extra runtime.
Closes https://github.com/espressif/esp-idf/issues/5301
This commit removes the following critical nested macros as follows:
- portENTER_CRITICAL_NESTED()
- portEXIT_CRITICAL_NESTED()
They are replaced with portSET_INTERRUPT_MASK_FROM_ISR() and
portCLEAR_INTERRUPT_MASK_FROM_ISR() which are the proper FreeRTOS interfaces.
Created a portmacro_deprecated.h for each port to contain deprecated API
that were originally from portmacro.h
- port.c and portmacro.h contents are now logically grouped
- Doxygen API descriptions added
- Added refactor todo markers
- Removed portmacro_priv.h. Contents now in portmacro.h
- Formatted files
Since dd849ffc, _rodata_start label has been moved to a different
linker output section from where the TLS templates (.tdata, .tbss)
are located. Since link-time addresses of thread-local variables are
calculated relative to the section start address, this resulted in
incorrect calculation of THREADPTR/$tp registers.
Fix by introducing new linker label, _flash_rodata_start, which points
to the .flash.rodata output section where TLS variables are located,
and use it when calculating THREADPTR/$tp.
Also remove the hardcoded rodata section alignment for Xtensa targets.
Alignment of rodata can be affected by the user application, which is
the issue dd849ffc was fixing. To accommodate any possible alignment,
save it in a linker label (_flash_rodata_align) and then use when
calculating THREADPTR. Note that this is not required on RISC-V, since
this target doesn't use TPOFF.
Enable shared stack watchpoint for overflow detection
Enable unit tests:
* "test printf using shared buffer stack" for C3
* "Test vTaskDelayUntil" for S2
* "UART can do poll()" for C3
