Solved
Inference Blocking on LL_ATON_OSAL_WFE() with ThreadX
Hello,
I am working on implementing an object detection application for the STM32N6570-DK, starting from the reference projects:
- STMicroelectronics/x-cube-n6-ai-people-detection-tracking
- STMicroelectronics/STM32N6-GettingStarted-ObjectDetection
I'm encountering a blocking issue when implementing the inference pipeline:
During debugging, I found that the inference execution blocks indefinitely at LL_ATON_OSAL_WFE() on line 105 of Appli\App\Src\app_nn.c. The code is stuck in the following loop:
static void NN_RunInference(void) {
LL_ATON_RT_RetValues_t ret;
do {
ret = LL_ATON_RT_RunEpochBlock(&NN_Instance_od_yolo_x_person);
if (ret == LL_ATON_RT_WFE) {
LL_ATON_OSAL_WFE(); // <-- Blocks here
}
} while (ret != LL_ATON_RT_DONE);
LL_ATON_RT_Reset_Network(&NN_Instance_od_yolo_x_person);
}I've reviewed similar posts in the community
, but it didn't resolve my issue.
Questions:
What could be the issue that blocks inference indefinitely at `LL_ATON_OSAL_WFE()`, and how to fix it?
Source Code:
My current implementation is available at: https://github.com/Foahh/stm32n6-ai-camera.git
