How to do CUDA programming on Jetson Nano?

Robotics & Edge Computing Jetson & Embedded Systems Jetson Nano
1

Dear community,

I did some search on the topic, and I can see the nvcc in /usr/local/cuda/bin, but I am lost in the next step. Could anyone tell me what should I do to practice cuda programming on Jetson nano? Thanks for help!

2

Hi,

It’s recommended to start it from the CUDA sample.

$ /usr/local/cuda-10.0/bin/cuda-install-samples-10.0.sh .
$ cd NVIDIA_CUDA-10.0_Samples/
$ make

You can execute and check the sample in NVIDIA_CUDA-10.0_Samples/

Thanks.

1 Like
3

The following commands are added to /home/.bashrc or nvcc cannot be found

export PATH=/usr/local/cuda-10.0/bin${PATH:+:${PATH}}$
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

However, when I $ nvcc file.cu, i got :

asyncAPI.cu:29:10: fatal error: helper_cuda.h: No such file or directory
#include <helper_cuda.h>
^~~~~~~~~~~~~~~
compilation terminated.

In the makefile I do find :

Common includes and paths for CUDA

INCLUDES := -I…/…/common/inc
LIBRARIES :=

and I changed to my actual directory for INCLUDES : /home/jetbot/Documents/CUDA_Prj/NVIDIA_CUDA-10.0_Samples/common/inc and I compile again use make. But the same issue is still there, could anyone help on this problem? Thanks.

4

Hi,

Makefile is used only when you compile the app by make.
It won’t be refereed if you use the compiler directly. Ex, $gcc …, $nvcc …

So there are two options for you.

1. Using the Makefile. Just type make:

$ make

2. Compile it with nvcc directly. So you will need to feed the header directory into the compiler:

$ nvcc deviceQuery.cpp -o test -I/home/nvidia/NVIDIA_CUDA-10.0_Samples/common/inc/

Thanks.