Nvcr.io/nvidia/l4t-ml:r32.4.3-py3 images NOT have some tensorrt's files

Robotics & Edge Computing Jetson & Embedded Systems Jetson AGX Xavier
1

platform:

  • NVIDIA Jetson AGX Xavier [16GB]
    • Jetpack 4.4 [L4T 32.4.3]
    • NV Power Mode: MAXN - Type: 0
      I pull a docker images:
      sudo docker pull nvcr.io/nvidia/l4t-ml:r32.4.3-py3
      then I create a container named “v10”
      sudo docker run -itd --name v10 nvcr.io/nvidia/l4t-ml:r32.4.3-py3
      when I enter this container,
      in the path: /usr/src/tensorrt
      this path have nothing, WHY.
      when I use sudo docker run -it --rm --runtime nvidia --network host nvcr.io/nvidia/l4t-ml:r32.4.3-py3
      in the path: /usr/src/tensorrt
      this path have filenames:
      bin data and samples
      why
2

Hi @wlmtobeno1, on JetPack 4.x, the containers mount CUDA/cuDNN/TensorRT from the host device when --runtime nvidia is used. You need --runtime nvidia to use the GPU.

On JetPack 5.x, CUDA/cuDNN/TensorRT are installed inside the container, but you should still use --runtime nvidia

5

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.