tegra-nvjpg print endlessly

Issue Overview

Users have reported an issue with the Nvidia Jetson Orin Nano Dev board, specifically encountering endless debug prints from the tegra-nvjpg module after flashing the r36.2 version to a custom board. The error message displayed is:

[  350.198195] tegra-nvjpg 15380000.nvjpg: _opp_is_duplicate: duplicate OPPs detected. Existing: freq: 115200000, volt: 0, enabled: 1. New: freq: 115200000, volt: 0, enabled: 1

Symptoms and Context

• Symptoms: Endless logging of duplicate OPP (Operating Performance Point) messages.
• Context: This issue arises after flashing the board with a specific configuration file and custom device tree modifications.
• Hardware/Software Specifications:
  • Flashing configuration includes various device tree files and settings for PCIe devices.
  • The issue occurs inconsistently; sometimes the board flashes successfully, while other times it fails.
• Frequency: The problem appears consistently after certain configurations are applied.
• Impact: The endless prints can clutter logs and may hinder performance or debugging efforts.

Possible Causes

• Hardware Incompatibilities: Custom hardware configurations may not align with expected parameters for the tegra-nvjpg module.

• Software Bugs: The r36.2 version may contain bugs affecting the handling of OPPs, as indicated by user experiences with earlier versions (r35.3.1) that do not exhibit this issue.

• Configuration Errors: Incorrect settings in the device tree or flashing configuration could lead to conflicts in how OPPs are defined or recognized.

• Driver Issues: Potential incompatibilities or bugs in the drivers associated with the tegra-nvjpg module could cause it to misreport duplicate OPPs.

• User Errors: Misconfigurations during setup or flashing processes might lead to unexpected behavior.

Troubleshooting Steps, Solutions & Fixes

Step-by-Step Diagnosis

1. Check Device Tree Configuration:

   • Review changes made to the device tree files to ensure they align with Nvidia’s specifications.
   • Verify that no unnecessary nodes are enabled in the device tree that could conflict with OPP settings.

2. Test Different Versions:

   • Flash earlier versions of the software (e.g., r35.3.1) to determine if the issue persists.
   • If successful, this may indicate a regression in the newer version.

3. Remove PCIe Devices:

   • Disable PCIe devices in the device tree and check if this resolves the issue.
   • Use commands such as:

     echo 1 > /sys/bus/pci/devices/<device_id>/remove
     

   • Note that simply disabling may not remove them from lspci, so consider using /delete-node/ in the device tree.

4. Log Analysis:

   • Collect full logs for further analysis; check for other related errors that might provide insight into underlying issues.
   • Use logging tools or pastebin for sharing large log files.

5. Cross-check with DevKits:

   • If available, test with a standard Nvidia DevKit to see if it triggers similar errors, which can help isolate whether it’s a custom board issue.

Recommended Solutions

• If removing PCIe devices does not resolve the issue, consider applying patches or modifications suggested by community members who have faced similar problems.

• Ensure all relevant firmware and driver updates are applied as per Nvidia’s guidelines for Jetson devices.

Best Practices for Prevention

• Regularly check for updates from Nvidia regarding known issues with specific versions of software and hardware compatibility notes.

• Maintain backups of stable configurations before applying new updates or changes to avoid regressions.

Unresolved Aspects

While several users have attempted various solutions, the root cause remains unclear, particularly whether it is a software bug in r36.2 or an incompatibility with custom hardware configurations. Further investigation may be required based on user feedback and additional testing results.