building with tensor rt 5.1.2 , cuda 10.0 ,cudnn 7.6.4 on Titan Rtx 24gb

AI & Data Science Deep Learning (Training & Inference) TensorRT
1

error:

ERROR: cuda/cudaConvolutionLayer.cpp (169) - Cudnn Error in execute: 8 (CUDNN_STATUS_EXECUTION_FAILED)
ERROR: cuda/cudaConvolutionLayer.cpp (169) - Cudnn Error in execute: 8 (CUDNN_STATUS_EXECUTION_FAILED)
Signal: SIGSEGV (Segmentation fault)

if i build with Quadro P4000, i am not getting this error.

please help me to solve this issue

2

Hello,

Recommend trying NVIDIA GPU Cloud (NGC) tensorrt optimized containers, which removes many of the host-side dependencies.
https://ngc.nvidia.com/signin

Thanks