Score:3

nvidia-cuda-toolkit ? not compatible with nvidia-495 driver

cf flag

When trying to install nvidia-cuda-toolkit (on ubuntu 21.10 with the nvidia-495 diver) I am asked to remove the nvidia driver. Why?

~> sudo apt install nvidia-cuda-toolkit
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
The following packages were automatically installed and are no longer required:
  dkms libnvidia-cfg1-495 libnvidia-fbc1-495 libnvidia-fbc1-495:i386
  libnvidia-gl-495 libxnvctrl0 nvidia-dkms-495 nvidia-kernel-common-495
  nvidia-prime nvidia-settings screen-resolution-extra
  xserver-xorg-video-nvidia-495
Use 'sudo apt autoremove' to remove them.
The following additional packages will be installed:
  g++-10 libaccinj64-11.3 libcub-dev libcublas11 libcublaslt11 libcudart11.0
  libcufft10 libcufftw10 libcuinj64-11.3 libcupti-dev libcupti-doc
  libcupti11.3 libcurand10 libcusolver11 libcusolvermg11 libcusparse11
  libegl-dev libgl-dev libgl1-mesa-dev libgles-dev libgles1 libglvnd-core-dev
  libglvnd-dev libglx-dev libjs-sphinxdoc libjs-underscore libnppc11
  libnppial11 libnppicc11 libnppidei11 libnppif11 libnppig11 libnppim11
  libnppist11 libnppisu11 libnppitc11 libnpps11 libnvblas11
  libnvidia-compute-470 libnvidia-ml-dev libnvjpeg11 libnvrtc-builtins11.3
  libnvrtc11.2 libnvtoolsext1 libnvvm4 libopengl-dev libopengl0
  libpthread-stubs0-dev libstdc++-10-dev libthrust-dev libvdpau-dev libx11-dev
  libxau-dev libxcb1-dev libxdmcp-dev node-html5shiv nsight-compute
  nsight-compute-target nsight-systems nsight-systems-target nvidia-cuda-dev
  nvidia-cuda-gdb nvidia-cuda-toolkit-doc nvidia-opencl-dev nvidia-profiler
  nvidia-visual-profiler ocl-icd-opencl-dev opencl-c-headers
  opencl-clhpp-headers x11proto-dev xorg-sgml-doctools xtrans-dev
Suggested packages:
  g++-10-multilib gcc-10-doc libstdc++-10-doc libvdpau-doc libx11-doc
  libxcb-doc opencl-clhpp-headers-doc
Recommended packages:
  libnvcuvid1
The following packages will be REMOVED:
  libnvidia-compute-495 libnvidia-compute-495:i386 libnvidia-decode-495
  libnvidia-decode-495:i386 libnvidia-encode-495 libnvidia-encode-495:i386
  nvidia-compute-utils-495 nvidia-driver-495 nvidia-utils-495
The following NEW packages will be installed:
  g++-10 libaccinj64-11.3 libcub-dev libcublas11 libcublaslt11 libcudart11.0
  libcufft10 libcufftw10 libcuinj64-11.3 libcupti-dev libcupti-doc
  libcupti11.3 libcurand10 libcusolver11 libcusolvermg11 libcusparse11
  libegl-dev libgl-dev libgl1-mesa-dev libgles-dev libgles1 libglvnd-core-dev
  libglvnd-dev libglx-dev libjs-sphinxdoc libjs-underscore libnppc11
  libnppial11 libnppicc11 libnppidei11 libnppif11 libnppig11 libnppim11
  libnppist11 libnppisu11 libnppitc11 libnpps11 libnvblas11
  libnvidia-compute-470 libnvidia-ml-dev libnvjpeg11 libnvrtc-builtins11.3
  libnvrtc11.2 libnvtoolsext1 libnvvm4 libopengl-dev libopengl0
  libpthread-stubs0-dev libstdc++-10-dev libthrust-dev libvdpau-dev libx11-dev
  libxau-dev libxcb1-dev libxdmcp-dev node-html5shiv nsight-compute
  nsight-compute-target nsight-systems nsight-systems-target nvidia-cuda-dev
  nvidia-cuda-gdb nvidia-cuda-toolkit nvidia-cuda-toolkit-doc
  nvidia-opencl-dev nvidia-profiler nvidia-visual-profiler ocl-icd-opencl-dev
  opencl-c-headers opencl-clhpp-headers x11proto-dev xorg-sgml-doctools
  xtrans-dev
0 upgraded, 73 newly installed, 9 to remove and 0 not upgraded.
Need to get 1,829 MB of archives.
After this operation, 4,523 MB of additional disk space will be used.
Do you want to continue? [Y/n]
cc flag
I doubt the CUDA code cares, but the deb packages were set up with some really unfortunate dependencies. See Answer 1077061 for a description of using Intel's .run script for the installation, allowing you to: 1)select your location instead of a system area,2)skip the Nvidia driver install.
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.