I’ve installed cudnn (version 6.0) via deb as per the instructions in the docs at http://docs.nvidia.com
But, after compiling and running mnistCUDNN I get the following information:
cudnnGetVersion() : 6021 , CUDNN_VERSION from cudnn.h : 6021 (6.0.21)
Host compiler version : GCC 4.9.3
There are 1 CUDA capable devices on your machine :
device 0 : sms 10 Capabilities 6.1, SmClock 1784.5 Mhz, MemSize (Mb) 6071, MemClock
4004.0 Mhz, Ecc=0, boardGroupID=0
*** stack smashing detected ***: ./mnistCUDNN terminated
Aborted (core dumped)
My system is Ubuntu 16.04.2
I’ve already installed Cuda 8.0 via deb as per the instructions in the docs at http://docs.nvidia.com
gcc version 4.9.3
NVIDIA UNIX x86_64 Kernel Module 384.90
Below are the relevant additions to .bashrc
export LD_LIBRARY_PATH=“$LD_LIBRARY_PATH:/usr/local/cuda-8.0/lib64:/usr/local/cuda-8.0/extras/CUPTI/lib64”
export CUDA_ROOT=/usr/local/cuda-8.0/
export CUDA_HOME=/usr/local/cuda-8.0
export LD_LIBRARY_PATH=“$LD_LIBRARY_PATH:/usr/local/cuda-8.0/lib64”
PATH=${CUDA_HOME}/bin:${PATH}
export PATH
export PATH=/usr/local/cuda-8.0/bin${PATH:+:${PATH}}
I’ve tried reinstalling cuda/tensorflow. Any help that you can provide would be appreciated.
Thanks.