Cannot open display: [TensorRT] ERROR: 1: [convolutionRunner.cpp::executeConv::458] Error Code 1: Cudnn (CUDNN_STATUS_EXECUTION_FAILED)

Hi, I am using a Jetson Nano for object recognition.
My model is a MobileNet v1 and I want to test it with a .mp4 video instead of the camera.

I got this error message. What is the problem and how can I solve this?
Thanks.

Cuda 10.2
cudnn 8.0.1
TensorRT 8.0.1.6

(dst:17245): Gtk-WARNING **: 05:46:51.267: cannot open display: 
[TensorRT] ERROR: 1: [convolutionRunner.cpp::executeConv::458] Error Code 1: Cudnn (CUDNN_STATUS_EXECUTION_FAILED)
[TensorRT] ERROR: 1: [convolutionRunner.cpp::executeConv::458] Error Code 1: Cudnn (CUDNN_STATUS_EXECUTION_FAILED)
[TensorRT] ERROR: 1: [convolutionRunner.cpp::executeConv::458] Error Code 1: Cudnn (CUDNN_STATUS_EXECUTION_FAILED)
[TensorRT] ERROR: 1: [convolutionRunner.cpp::executeConv::458] Error Code 1: Cudnn (CUDNN_STATUS_EXECUTION_FAILED)
[TensorRT] ERROR: 1: [convolutionRunner.cpp::executeConv::458] Error Code 1: Cudnn (CUDNN_STATUS_EXECUTION_FAILED)
[TensorRT] ERROR: 1: [convolutionRunner.cpp::executeConv::458] Error Code 1: Cudnn (CUDNN_STATUS_EXECUTION_FAILED)
[TensorRT] ERROR: 1: [convolutionRunner.cpp::executeConv::458] Error Code 1: Cudnn (CUDNN_STATUS_EXECUTION_FAILED)
Segmentation fault (core dumped)

Hi,

(dst:17245): Gtk-WARNING **: 05:46:51.267: cannot open display: 

Based above, do you have a physical display connected?
If yes, please try to export the below global variable to see if it helps.

$ export DISPLAY=:0   # 1 for some usecase

Thanks.

Hi, actually I don’t have a physical display, so I removed the lines of codes that were related to the display and it is working now.

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