[root@localhost ~]# nvidia-smi NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running. [root@localhost ~]# lsmod | grep nvidia [root@localhost ~]# [root@localhost ~]# rpm -qa |grep nvidia nvidia-driver-NVML-530.30.02-4.an23.x86_64 nvidia-driver-libs-530.30.02-4.an23.x86_64 nvidia-driver-cuda-libs-530.30.02-4.an23.x86_64 nvidia-driver-NvFBCOpenGL-530.30.02-4.an23.x86_64 nvidia-persistenced-530.30.02-4.an23.x86_64 nvidia-driver-cuda-530.30.02-4.an23.x86_64 nvidia-driver-devel-530.30.02-4.an23.x86_64 nvidia-driver-530.30.02-4.an23.x86_64 nvidia-kmod-common-530.30.02-4.an23.x86_64 kmod-nvidia-latest-dkms-530.30.02-4.an23.x86_64 nvidia-modprobe-530.30.02-4.an23.x86_64 nvidia-settings-530.30.02-4.an23.x86_64 nvidia-xconfig-530.30.02-4.an23.x86_64 [root@localhost ~]# rpm -qa |grep cuda cuda-toolkit-config-common-12.1.105-4.an23.noarch cuda-toolkit-12-config-common-12.1.105-4.an23.noarch cuda-toolkit-12-1-config-common-12.1.105-4.an23.noarch cuda-nvml-devel-12-1-12.1.105-4.an23.x86_64 cuda-driver-devel-12-1-12.1.105-4.an23.x86_64 cuda-cudart-12-1-12.1.105-4.an23.x86_64 cuda-opencl-12-1-12.1.105-4.an23.x86_64 cuda-nvrtc-12-1-12.1.105-4.an23.x86_64 cuda-nvdisasm-12-1-12.1.105-4.an23.x86_64 cuda-cupti-12-1-12.1.105-4.an23.x86_64 cuda-nvprof-12-1-12.1.105-4.an23.x86_64 cuda-cccl-12-1-12.1.109-4.an23.x86_64 cuda-libraries-12-1-12.1.1-4.an23.x86_64 cuda-cudart-devel-12-1-12.1.105-4.an23.x86_64 cuda-nvrtc-devel-12-1-12.1.105-4.an23.x86_64 cuda-opencl-devel-12-1-12.1.105-4.an23.x86_64 cuda-nsight-compute-12-1-12.1.1-4.an23.x86_64 cuda-profiler-api-12-1-12.1.105-4.an23.x86_64 cuda-libraries-devel-12-1-12.1.1-4.an23.x86_64 cuda-nvtx-12-1-12.1.105-4.an23.x86_64 cuda-nvprune-12-1-12.1.105-4.an23.x86_64 cuda-nvcc-12-1-12.1.105-4.an23.x86_64 cuda-gdb-12-1-12.1.105-4.an23.x86_64 cuda-documentation-12-1-12.1.105-4.an23.x86_64 cuda-cuxxfilt-12-1-12.1.105-4.an23.x86_64 cuda-cuobjdump-12-1-12.1.111-4.an23.x86_64 cuda-compiler-12-1-12.1.1-4.an23.x86_64 cuda-nsight-systems-12-1-12.1.1-4.an23.x86_64 cuda-nsight-12-1-12.1.105-4.an23.x86_64 cuda-nvvp-12-1-12.1.105-4.an23.x86_64 cuda-visual-tools-12-1-12.1.1-4.an23.x86_64 cuda-demo-suite-12-1-12.1.105-4.an23.x86_64 nvidia-driver-cuda-libs-530.30.02-4.an23.x86_64 nvidia-driver-cuda-530.30.02-4.an23.x86_64 cuda-drivers-530.30.02-4.an23.x86_64 cuda-runtime-12-1-12.1.1-4.an23.x86_64 cuda-sanitizer-12-1-12.1.105-4.an23.x86_64 cuda-command-line-tools-12-1-12.1.1-4.an23.x86_64 cuda-tools-12-1-12.1.1-4.an23.x86_64 cuda-toolkit-12-1-12.1.1-4.an23.x86_64 cuda-12-1-12.1.1-4.an23.x86_64 cuda-12.1.1-4.an23.x86_64
安装环境的kernel与kernel-devel版本不匹配导致nvidia-kmod安装史dkms执行失败。 考虑在nvidia-kmod安装时dkms执行失败给出提示。 另外由于kernel-devel包允许存在多版本,dkms构建时依赖当前内核版本,仍旧需要给出指导文档来说明如何正确安装nvidia-driver。
已在kmod-nvidia-latest-dkms安装时增加检测,安装时如若OS上未安装kernel-devel-$(uname -r),则会给出提示。 类似于如下提示: Running scriptlet: kmod-nvidia-latest-dkms-3:470.199.02-3.an8.x86_64 32/34 You should install kernel-devel-5.10.134-14.1.an8.x86_64 first to build nvidia module for current kernel. warning: %post(kmod-nvidia-latest-dkms-3:470.199.02-3.an8.x86_64) scriptlet failed, exit status 1 Error in POSTIN scriptlet in rpm package kmod-nvidia-latest-dkms