Installing CUDA on Ubuntu 23.10 – libt5info not installable
Today I’ve installed from scratch Ubuntu 23.10 on my computer.
After having installed all the software I need, I tried to install CUDA from the NVIDIA website, following their instructions: https://developer.nvidia.com/cuda-downloads?target_os=Linux&target_arch=x86_64&Distribution=Ubuntu&target_version=22.04&target_type=deb_local
There’s written that the latest version supported is 22.04 anyway I tried…, the problem happens with the last command
sudo apt-get -y install cuda-toolkit-12-3 because the terminal gives me an error:
Reading package lists... Done Building dependency tree... Done Reading state information... Done Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation: The following packages have unmet dependencies: nsight-systems-2023.3.3 : Depends: libtinfo5 but it is not installable E: Unable to correct problems, you have held broken packages.
I tried to manually install such library but I get
E: Package 'libtinfo5' has no installation candidate
Is there any way to install cuda on this system ?
I was able to install manually the package needed from the Ubuntu 23.04 (lunar) repo: https://packages.ubuntu.com/lunar/libtinfo5. Then, running
sudo apt install cuda resulted in no errors and CUDA was correctly installed.
libtinfo5 package isn’t available in Ubuntu 23.10’s default repositories yet. We can install it by adding the
universe repo for Ubuntu 23.04 (Lunar Lobster).
Open a terminal window and run:
sudo nano /etc/apt/sources.list
Add this line (adds the Ubuntu 23.04 aka "Lunar Lobster" universe repository to
deb http://archive.ubuntu.com/ubuntu/ lunar universe
Save and exit, then run:
sudo apt update
…and now the install command for CUDA should work, automatically downloading and installing
libtinfo5 while installing CUDA.