Orin Nano DP output not normal

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin Nano
,
1

Hi,we developed a carrier board using the Orin Nano module, and the SOC DP signal is sent to the video processing chip LT8712SX-U2. The LT8712SX-U2 chip is divided into two channels, one channel is HDMI output, and the other channel is EDP signal used to drive the EDP LCD screen. When the device is just turned on, there is a picture output from HDMI, but when it is about to enter the Ubuntu login interface, there is no picture. However, it can be seen that the monitor received the HDMI signal and did not decode and display it properly. Through the log, we can see an error message,and the error log is as follows:
sudo cat /var/lib/gdm3/.local/share/xorg/Xorg.0.log | grep EE
[ 16.769] Current Operating System: Linux nvidia-desktop 5.15.148-tegra #4 SMP PREEMPT Thu Apr 3 17:38:00 CST 2025 aarch64
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[ 16.843] (EE) NVIDIA(0): Failed to allocate NVIDIA Error Handler
[ 16.896] (II) Initializing extension MIT-SCREEN-SAVER.
Can you help analyze how the error came about? How can DP output be normal?

2

This is normal, and can be ignored:

[ 16.843] (EE) NVIDIA(0): Failed to allocate NVIDIA Error Handler
[ 16.896] (II) Initializing extension MIT-SCREEN-SAVER.

You need a serial console boot log. See:
https://jetsonhacks.com/2019/04/19/jetson-nano-serial-console/

There is also a chance that you can use ssh to log in. If you can get a login via either ssh or serial console, or if you can get in to text mode via “CTRL-ALT-F2”, then you can get the Xorg log. The relevant log:

cd /var/log
ls -ltr Xorg.*.log | tail -n 1

See if you can get access to that log.

For custom carrier boards there is often an issue with device tree which needs to change to support something related to one of these:

  • Hot plug detect.
  • i2c bus power related to EDID (the GPU powers the i2c circuitry in the monitor; even if the GPU knows there is a monitor plug-in event, then it can’t find out more unless it provides that i2c power).
  • The DDC wire for EDID data.

Between serial console boot log and that Xorg log you can find a lot more information. Odds are in favor of one of those device tree settings being at issue.