最近在T440笔记本的ubuntu 18.04系统上安装最新的CUDA Toolkit 10.1 Update 2
之后,发现 systemd-udevd
进程CPU占用特别高,执行 sudo /lib/systemd/systemd-udevd -D
,会发现持续输出如下信息:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 |
RUN '/bin/systemctl start --no-block nvidia-persistenced.service' /lib/udev/rules.d/71-nvidia.rules:12 RUN '/sbin/modprobe nvidia-modeset' /lib/udev/rules.d/71-nvidia.rules:16 RUN '/sbin/modprobe nvidia-drm' /lib/udev/rules.d/71-nvidia.rules:20 RUN '/sbin/modprobe nvidia-uvm' /lib/udev/rules.d/71-nvidia.rules:24 RUN '/usr/bin/nvidia-smi' /lib/udev/rules.d/71-nvidia.rules:28 starting '/bin/systemctl start --no-block nvidia-persistenced.service' Process '/bin/systemctl start --no-block nvidia-persistenced.service' succeeded. starting '/sbin/modprobe nvidia-modeset' seq 115679 queued, 'remove' 'module' seq 115680 queued, 'add' 'module' seq 115681 queued, 'add' 'slab' seq 115682 queued, 'add' 'drivers' seq 115681 running seq 115682 running seq 115681 processed seq 115683 queued, 'remove' 'slab' seq 115684 queued, 'remove' 'drivers' seq 115683 running seq 115683 processed seq 115682 processed seq 115684 running seq 115684 processed seq 115685 queued, 'remove' 'module' '/sbin/modprobe nvidia-modeset'(err) 'modprobe: ERROR: could not insert 'nvidia_modeset': No such device' Process '/sbin/modprobe nvidia-modeset' failed with exit code 1. |
解决方法如下:
1 |
$ sudo rm -rf /lib/udev/rules.d/71-nvidia.rules |