Supporting MIPI Camera with 8-bit and 16-bit Formats Using a Single Device Tree

Issue Overview

Users are experiencing difficulties configuring a custom MIPI camera to support both 8-bit and 16-bit per pixel imaging formats using a single device tree on the Nvidia Jetson Orin Nano Dev board. The 16-bit format requires 4 MIPI lanes, while the 8-bit format only needs two lanes. Currently, separate device trees are needed to capture both formats, as the "bus-width" parameter must be set to 4 for 4 lanes and 2 for 2 lanes. Users are seeking a solution to support both formats with a single device tree, either through programmatic means via the camera driver or by confining the format differences to the sensor "modes".

Possible Causes

1. Device Tree Configuration Limitations: The current device tree structure may not support dynamic lane configuration based on the pixel format.

2. Driver Inflexibility: The camera driver might not be designed to handle dynamic lane switching based on the selected pixel format.

3. Hardware Constraints: The MIPI interface on the Jetson Orin Nano Dev board may have limitations in dynamically switching between 2-lane and 4-lane configurations.

4. Sensor Mode Configuration: The sensor modes in the device tree might not be properly defined to accommodate both 8-bit and 16-bit formats with their respective lane requirements.

Troubleshooting Steps, Solutions & Fixes

1. Modify Sensor Modes:

   • Define both 2-lane and 4-lane settings within different sensor modes in the device tree.
   • Add the use_sensor_mode_id property to the device tree to enable mode-specific configurations.
   • Example command for mode selection:

     v4l2-ctl -d /dev/video0 --set-fmt-video=width=1280,height=720,pixelformat=RG10 --set-ctrl bypass_mode=0 --set-ctrl sensor_mode=4 --stream-mmap --stream-count=100
     

2. Adjust Pixel Clock Rate:

   • For the 8-bit format, consider revising it as a 4-lane setting by lowering the pixel clock rate. This approach may allow you to maintain a consistent 4-lane configuration in the device tree while supporting both 8-bit and 16-bit formats.

3. Investigate Driver Modifications:

   • If the above solutions don’t work, consider modifying the camera driver to dynamically adjust the lane configuration based on the selected pixel format.
   • This would require expertise in driver development and a deep understanding of the Jetson Orin Nano’s camera subsystem.

4. Review Port Definitions:

   • Examine the port definitions in the device tree, particularly the bus-width parameter.
   • Current example of port definition:

     tegra-capture-vi  {
         num-channels = <2>;
         ports {
             #address-cells = <1>;
             #size-cells = <0>;
             vi_port0: port@0 {
                 reg = <0>;
                 custom_cam_vi_in0: endpoint {
                     port-index = <1>;
                     bus-width = <4>;
                     remote-endpoint = <&custom_cam_csi_out0>;
                 };
             };
             port@1 {
                 reg = <1>;
                 custom_cam_vi_in1: endpoint {
                     port-index = <2>;
                     bus-width = <4>;
                     remote-endpoint = <&custom_cam_csi_out1>;
                 };
             };
         };
     };
     

   • Investigate if it’s possible to make the bus-width parameter dynamic or mode-specific.

5. Consult Nvidia Documentation:

   • Review the Jetson Linux Developer Guide, particularly the section on Device Properties, for any advanced configuration options that might help address this issue.
   • Look for any application notes or errata related to MIPI camera configurations on the Jetson Orin Nano.

6. Community Support:

   • If the issue persists, consider reaching out to the Nvidia Developer Forums or the Jetson community for more specialized assistance.
   • Provide detailed information about your camera model, exact Jetson Orin Nano version, and the steps you’ve already taken to troubleshoot the problem.

7. Hardware Verification:

   • Ensure that your custom MIPI camera is functioning correctly and that the issue is not caused by a hardware defect.
   • Test the camera with other devices or development boards, if possible, to isolate whether the problem is specific to the Jetson Orin Nano configuration.