This commit fixes a potential issue of illegal memory access when
accessing the various task lists in FreeRTOS while fetching tasks using
the xTaskGetNext() function.
- move .tbss to NOLOAD section
- remove xtensa-specific entities from riscv scripts
- explicit eh_frame terminator instead of "align magic"
- 80 characters line length limit
- refactor comments
- discard .rela sections (the rela data will go to relates sections)
This commit updates the dependency for
CONFIG_FREERTOS_TASK_FUNCTION_WRAPPER to be dependent explicitly on
CONFIG_ESP_SYSTEM_PANIC_GDBSTUB or CONFIG_ESP_SYSTEM_GDBSTUB_RUNTIME
instead of CONFIG_ESP_GDBSTUB_ENABLED which caused FreeRTOS Task
Function Wrappers to be enabled even if the esp_gdbstub component was
part of the build.
- vTaskPreemptionDisable()/vTaskPreemptionEnable() are no longer available in
single-core builds.
- xTaskIncrementTick() calls must now be wrapped by critical section
- Minimal Idle Task renamed to Passive Idle Task
- Port critical section APIs updated
fix(riscv): Updated RISC-V functions to set interrupt threshold for CLIC targets
Closes IDFCI-2033, IDFCI-2034, IDF-8090, and IDF-8117
See merge request espressif/esp-idf!29055
This commit introduce SOC_MEM_NON_CONTIGUOUS_SRAM flag (that enebled for
esp32p4). If SOC_MEM_NON_CONTIGUOUS_SRAM is enabled:
- LDFLAGS+=--enable-non-contiguous-regions
- ldgen.py replaces "arrays[*]" from sections.ld.in with objects under
SURROUND keyword. (e.g. from linker.lf: data -> dram0_data SURROUND(foo))
- "mapping[*]" - refers to all other data
If SOC_MEM_NON_CONTIGUOUS_SRAM, sections.ld.in file should contain at
least one block of code like this (otherwise it does not make sense):
.dram0.bss (NOLOAD) :
{
arrays[dram0_bss]
mapping[dram0_bss]
} > sram_low
.dram1.bss (NOLOAD) :
{
/* do not place here arrays[dram0_bss] because it may be splited
* between segments */
mapping[dram0_bss]
} > sram_high
This commit added the RISC-V utility functions to set the interrupt
threshold for CLIC targets by using direct register value writes.
This makes the functions more efficient during run-time.
This is done to improve the critical section enter and exit performance on esp32p4.
This commit replaces the use of portNUM_PROCESSORS and configNUM_CORES
macros in all of ESP-IDF. These macros are needed to realize an SMP
scenario by fetching the number of active cores FreeRTOS is running on.
Instead, a new Kconfig option, CONFIG_FREERTOS_NUMBER_OF_CORES, has been
added as a proxy for the FreeRTOS config option, configNUMBER_OF_CORES.
This new commit is now used to realize an SMP scenario in various places
in ESP-IDF.
[Sudeep Mohanty: Added new Kconfig option CONFIG_FREERTOS_NUMBER_OF_CORES]
Signed-off-by: Sudeep Mohanty <sudeep.mohanty@espressif.com>
This commit changes the approach used to override select function
in FreeRTOS simulator.
The previous approach relied on '--wrap', which is a feature of GNU
linker that performs symbol substitution at link time. Sadly this
feature is not present in macOS linker.
The alternative solution is to define 'select' wrapper as a regular
(exported) symbol, preventing it from being loaded from system
libraries. To call the "real" select implementation we use dlsym
function and find the function pointer at run time. This solution
works on both Linux and macOS.
This commit adds esp_backtrace_print_all_tasks() which prints the backtraces
of all tasks at runtime.
Closes https://github.com/espressif/esp-idf/issues/9708
CLoses https://github.com/espressif/esp-idf/pull/11575
[Omar Chebib: Prevent task switching while printing backtraces of tasks.]
[Omar Chebib: Ensure all task stacks are flushed from register to RAM.]
[Omar Chebib: Removed esp_task_snapshot_to_backtrace_frame() as task snapshot is private API.]
[Omar Chebib: Added test case for esp_backtrace_print_all_tasks().]
Signed-off-by: Omar Chebib <omar.chebib@espressif.com>
IDF FreeRTOS v10.5.1 no longer accepts out of range xCoreID arguments in
"PinnedToCore" task creation functions when building for single-core. This
commit fixes those violations through ESP-IDF.
This commit does the following:
- removes the xCoreID member from the TCB when building for single-core
- xCoreID is no longer hard set to 0 when calling "PinnedToCore" task creation
functions in single-core
- Tidy up or add missing xCoreID asserts for functions that take xCoreID as an
argument:
- Functions that set/query a variable of a particular core will call
taskVALID_CORE_ID() to ensure ) 0 <= xCoreID < configNUMBER_OF_CORES
- Task creation functions that accept xCoreID also call taskVALID_CORE_ID()
but also allow tskNO_AFFINITY.
- Fix TaskStatus_t
- Remove xCoreID from TaskStatus_t if configTASKLIST_INCLUDE_COREID is not
defined.
- Set xCoreID to 0 when calling vTaskGetInfo() in single-core builds
This commit refactors the following macros so that calling them no longer
requires referencing pxTCB->xCoreID.
- taskIS_YIELD_REQUIRED()
- taskIS_YIELD_REQUIRED_USING_PRIORITY()
- taskIS_AFFINITY_COMPATIBLE()
This commit deprecates the "freertos/xtensa_context.h" and "xtensa/xtensa_context.h"
include paths. Users should use "xtensa_context.h" instead.
- Replace legacy include paths
- Removed some unnecessary includes of "xtensa_api.h"
- Add warning to compatibility header
This commit deprecates the "freertos/xtensa_api.h" and "xtensa/xtensa_api.h"
include paths. Users should use "xtensa_api.h" instead.
- Replace legacy include paths
- Removed some unnecessary includes of "xtensa_api.h"
- Replaced some calls with "esp_cpu_..." equivalents
- Add warning to compatibility header
This commit refactors suspend-resume tests by moving them to appropraite
test files. This commit also removes tests which are not necessary or
have been subsumed by refactored tests.
