🛠 Advanced First Boot of LattePanda Mu Ultra

LattePanda Mu Ultra: x86 Compute Module for Embedded AI · avance · Guide technique pour l'assemblage et le premier démarrage sous Linux avec validation des pilotes graphiques et réseau.

Objectives

This tutorial aims to assemble the LattePanda Mu Ultra module on its Mini Carrier Board and perform a first boot under Linux Ubuntu 24.04 or 26.04 LTS. The focus is on critical hardware validation: recognition of the Intel I226-V Ethernet controller, activation of the Intel Arc Graphics GPU via the xe driver, and verification of LPDDR5X memory. Unlike beginner approaches, we prioritize an optimized system configuration to exploit 115 TOPS of AI power and the 4.8 GHz frequency of the P-cores.

Required Materials

Steps

  1. Thermal Assembly: Apply a thin layer of thermal paste on the Mu Ultra processor die. Secure the aluminum heatsink, respecting recommended torque specs to avoid PCB flexing. Adequate cooling is crucial to maintain 4.8 GHz on Lion Cove cores.
  1. Insertion and Power: Gently insert the module into the Mini Carrier Board connector. Connect DC power. Plug the Ethernet cable into the RJ45 port (I226-V controller) and connect the screen and keyboard.
  1. BIOS Access: Power on and press Del or F2 to enter UEFI. Verify NVMe storage detection and that memory shows 16 GB. Enable VT-d support if needed for virtualization.
  1. Linux Installation: Boot from the USB drive. Choose "Try Ubuntu". Once the desktop loads, open a terminal to check drivers.
  1. Graphics Validation: The Arc GPU should be recognized by the xe driver. Check with:
lspci -k | grep -A 3 VGA

Ensure the xe driver is loaded. If OpenGL fails, install full mesa packages.

  1. Network Validation: Verify connection to the Intel I226-V controller:
ip addr show
ethtool eth0

The link should be active at 2.5 Gbps if the router supports it.

Verification

Run lscpu to confirm the presence of 8 cores (4 P-cores, 4 E-cores). Use free -h to validate 16 GB of LPDDR5X. Finally, run glxinfo | grep "OpenGL version" to confirm OpenGL 4.6 support by the Intel Arc GPU.

Further Reading

Configure the Linux kernel (7.0.0-38 or higher) to optimize thermal management of hybrid cores. Explore embedded AI APIs to exploit the available 115 TOPS.

Other tutorials for this board

Code examples for this board

← Back to the board 🛠 All tutorials