When I install NVIDIA drivers on Centos/Ubuntu by method described at following link
https://developer.nvidia.com/cuda-11.1.0-download-archive?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=2004&target_type=debnetwork
I see that nvidia.ko and other modules are successfully built and accessible under /var/lib/modules filesystem subtree. But when I perform instructions related to CentOS, i. e.
https://developer.nvidia.com/cuda-11.1.0-download-archive?target_os=Linux&target_arch=x86_64&target_distro=CentOS&target_version=7&target_type=rpmnetwork
https://developer.nvidia.com/cuda-11.1.0-download-archive?target_os=Linux&target_arch=x86_64&target_distro=CentOS&target_version=8&target_type=rpmnetwork
there are no *.ko (or compressed *.ko.xz) modules built and installed. So there are no nvidia devices under /dev, impossible to load NVIDIA drivers by insmod, and I see following lines in the syslog:
Dec 22 10:51:37 centos7gputest nvidia-persistenced: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 0 has read and write permissions for those files.
How can I install NVIDIA kernel modules on CentOS? Should I build them by hands? (configure & make etc. inside source directory like /usr/src/nvidia-460.27.04/) or there are other official methods?