自从上次更新 Nvidia 驱动程序后,我就无法再使用 GPU 了。Ubuntu 18.04.4 LTS。
通过检查/var/log/apt/history.log
文件,我发现:
Start-Date: 2020-06-26
06:30:51 Commandline: /usr/bin/unattended-upgrade
Upgrade: libnvidia-common-440:amd64 (440.82-0ubuntu0~0.18.04.2, 440.100-0ubuntu0.18.04.1)
End-Date: 2020-06-26 06:30:51
根据建议这里,我试过了sudo apt update
,sudo apt upgrade
并重启了机器。不幸的是,这没有帮助。
跑步nvidia-smi
让我:
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
运行lspci -k | grep -EA3 'VGA|3D|Display'
结果:
17:00.0 VGA compatible controller: NVIDIA Corporation GP102 [GeForce GTX 1080 Ti] (rev a1)
Subsystem: ASUSTeK Computer Inc. GP102 [GeForce GTX 1080 Ti]
Kernel modules: nvidiafb, nouveau
17:00.1 Audio device: NVIDIA Corporation GP102 HDMI Audio Controller (rev a1)
--
65:00.0 VGA compatible controller: NVIDIA Corporation GP102 [TITAN Xp] (rev a1)
Subsystem: NVIDIA Corporation GP102 [TITAN Xp]
Kernel modules: nvidiafb, nouveau
65:00.1 Audio device: NVIDIA Corporation GP102 HDMI Audio Controller (rev a1)
如果我运行 python 并且:
import tensorflow as tf
tf.config.get_visible_devices()
我有错误:
2020-06-27 17:39:04.117940: E tensorflow/stream_executor/cuda/cuda_driver.cc:313] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
2020-06-27 17:39:04.117989: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (imt_deep): /proc/driver/nvidia/version does not exist
知道如何修复这个问题吗?
谢谢你,G。
编辑:
的输出uname -a
为:
Linux imt_deep 4.19.0-041900-generic #201810221809 SMP Mon Oct 22 22:11:45 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux