Trouble installing CUDA Toolkit 9.0 using JetPack

Robotics & Edge Computing Jetson & Embedded Systems Jetson TX1
1

I’m trying to install CUDA Toolkit 9.0 on a system that has had NVIDIA rootfs 28.2 installed previously (a compatible version from JetPack3.3).

To do this, I’ve selected the ‘custom’ radio button in JetPack3.3, and then selected only ‘CUDA Toolkit’, then ‘install 9.0’ in the dropdown list.

I then click run, and XTERM pops up as usual. The cuda-repo-l4t-9-0-local_9.0.252-1_arm64.deb is pushed to the target, then there is a delay and an exit code ‘5’, as shown below:

1610×976 60.2 KB

The error seems to come from the program InstallUtil, which didn’t seem to respond to queries for help (-h, --help).

My host system is 16.04. There is enough free flash on the target system for CUDA

2

I’m not sure about which versions might being mixed, but usually software packages from one release to another are not compatible (there are probably some exceptions).

What cuda directories do you have in “/var/”? Should be:

/var/cuda-repo-9-0-local/

What do you see from:

dpkg -l | egrep 'cuda-repo-l4t-9-0-local'
3

@linuxdev

Thanks, here’s the output:

nvidia@tegra-ubuntu:~$ ls /var/cuda-repo-9-0-local/
7fa2af80.pub                                     cuda-libraries-9-0_9.0.252-1_arm64.deb
cuda-command-line-tools-9-0_9.0.252-1_arm64.deb  cuda-libraries-dev-9-0_9.0.252-1_arm64.deb
cuda-core-9-0_9.0.252-1_arm64.deb                cuda-license-9-0_9.0.252-1_arm64.deb
cuda-cublas-9-0_9.0.252-1_arm64.deb              cuda-minimal-build-9-0_9.0.252-1_arm64.deb
cuda-cublas-dev-9-0_9.0.252-1_arm64.deb          cuda-misc-headers-9-0_9.0.252-1_arm64.deb
cuda-cudart-9-0_9.0.252-1_arm64.deb              cuda-npp-9-0_9.0.252-1_arm64.deb
cuda-cudart-dev-9-0_9.0.252-1_arm64.deb          cuda-npp-dev-9-0_9.0.252-1_arm64.deb
cuda-cufft-9-0_9.0.252-1_arm64.deb               cuda-nvgraph-9-0_9.0.252-1_arm64.deb
cuda-cufft-dev-9-0_9.0.252-1_arm64.deb           cuda-nvgraph-dev-9-0_9.0.252-1_arm64.deb
cuda-curand-9-0_9.0.252-1_arm64.deb              cuda-nvml-dev-9-0_9.0.252-1_arm64.deb
cuda-curand-dev-9-0_9.0.252-1_arm64.deb          cuda-nvrtc-9-0_9.0.252-1_arm64.deb
cuda-cusolver-9-0_9.0.252-1_arm64.deb            cuda-nvrtc-dev-9-0_9.0.252-1_arm64.deb
cuda-cusolver-dev-9-0_9.0.252-1_arm64.deb        cuda-runtime-9-0_9.0.252-1_arm64.deb
cuda-cusparse-9-0_9.0.252-1_arm64.deb            cuda-samples-9-0_9.0.252-1_arm64.deb
cuda-cusparse-dev-9-0_9.0.252-1_arm64.deb        cuda-toolkit-9-0_9.0.252-1_arm64.deb
cuda-documentation-9-0_9.0.252-1_arm64.deb       Packages.gz
cuda-driver-dev-9-0_9.0.252-1_arm64.deb          Release
cuda-gdb-src-9-0_9.0.252-1_arm64.deb             Release.gpg


nvidia@tegra-ubuntu:~$ dpkg -l | egrep 'cuda-repo-l4t-9-0-local'
ii  cuda-repo-l4t-9-0-local               9.0.252-1                                     arm64        cuda repository configuration files
nvidia@tegra-ubuntu:~$
4

So the CUDA repo itself is installed. I’m not sure about the meaning of the exit code 5, but it looks like CUDA 9 was installed and should be available. Was there some other problem showing up which makes it appear CUDA is not there?

Note that you can do “apt search cuda-gdb” (or any package in that directory) and it should show up as available even if not currently installed. Then you can use “sudo apt-get install …package…” to actually have the package installed.

5

@linuxdev

Awesome, this solved the problem. Running “sudo apt install cuda-gdb-src-9-0” got me what I needed.

Thank you!

Mark