BUSIF1 ERR: 0x20 and Epoch Controller ERROR interrupt: EC_IRQ = 0x0000000a with STM32N6570-DK

[Firmin]

Hello everyone,

I am encountering errors when running Semantic Segmentation, Object Detection, and Pose Estimation models on my STM32N6570-DK board using STM32 Model Zoo Services. Here are the details of my setup and the errors I am facing:

Hardware and Versions:

• Board: STM32N6570-DK
• Models: Semantic Segmentation, Object Detection, Pose Estimation
• Tools: STM32Cube.AI 4.0.0, GCC 14.3.1, HAL 1.3.0, STMCubeIDE 2.1.1

I followed the README instructions from the STM32 Model Zoo Services repository to test the mentioned models. I compiled and flashed the firmware onto the STM32N6570-DK board.

Errors Encountered:

plaintext

Copy

Bus interface interrupt
BUSIF0 ERR: 0x0
BUSIF1 ERR: 0x20
Epoch Controller ERROR interrupt: EC_IRQ = 0x0000000a
Epoch Controller opcode counter: 0x00000000
Epoch Controller label: 0x00000000
ATON_STD_IRQHandler()@871: irqs=0xlx
assertion "0" failed: file "[...]/stm32ai-modelzoo-services/application_code/.../ll_aton_runtime.c", line 875, function: __LL_ATON_RT_IrqErr

Questions:

1. What could be causing the BUSIF1 ERR: 0x20 error?
2. Why is the Epoch Controller generating an error interrupt with EC_IRQ = 0x0000000a?
3. Are there specific configurations for Semantic Segmentation, Object Detection, and Pose Estimation models that could resolve these errors?
4. Are there known code examples or fixes for these errors?

Thank you in advance for your help!

[Julian E.]

Hi @Firmin,

I have never seen these errors.

Could you please describe with more details what you did exactly with the model zoo?

Did you use the deployment mode, or did you use the application code directly? (we advise using the deployment mode)

What model are you using? Please share your .yaml if using the deployment mode

It could be an issue with the new STMCubeIDE 2.1.1 version. 

In the readme, you should have a required/tested version. For object detection, I think it is 1.17.0. 

Maybe it comes from that.

Please also check that the ST Edge AI Core version required is also the 4.0.0, else it could mean that the applications versions are not the latest ones.

Have a good day,

Julian

[Firmin]

Hi @Julian E. 
I've used the deployment mode, STEdgeAI 4.0.0, yolov8n_256_quant_pc_uf_pose_coco-st_OE_3_3_1 as model.

my .yaml :

model:
   model_path: https://github.com/stm32-hotspot/ultralytics/raw/refs/heads/main/examples/YOLOv8-STEdgeAI/stedgeai_models/pose_estimation/yolov8n_256_quant_pc_uf_pose_coco-st.tflite
   model_type: yolo_mpe

operation_mode: deployment

dataset:
   class_names: [person]
   keypoints: 17  # 13
   dataset_name: coco

preprocessing:
   resizing:
      aspect_ratio: crop
      interpolation: nearest
   color_mode: rgb

postprocessing:
   kpts_conf_thresh: 0.15
   confidence_thresh: 0.5
   max_detection_boxes: 10
   NMS_thresh: 0.5

tools:
   stedgeai:
      optimization: balanced
      on_cloud: False
      path_to_stedgeai: /opt/ST/STEdgeAI/4.0/Utilities/linux/stedgeai
   path_to_cubeIDE: /opt/st/stm32cubeide_2.1.1/stm32cubeide
deployment:
   c_project_path: ../application_code/pose_estimation/STM32N6/
   IDE: GCC
   verbosity: 1
   hardware_setup:
      serie: STM32N6
      board: STM32N6570-DK

mlflow:
   uri: ./tf/src/experiments_outputs/mlruns

hydra:
   run:
      dir: ./tf/src/experiments_outputs/${now:%Y_%m_%d_%H_%M_%S}

 
Thank you in advance for your help.
Firmin.