WSL2 Ubuntu22.04下TensorFlow无法加载libnvinfer.so.7及GPU问题

问题背景 #

• WSL2 Ubuntu 22.04环境，按TensorFlow官方pip指南完成TensorFlow 2.11.0安装
• Windows端已安装NVIDIA驱动，其他WSL2实例可正常运行GPU仿真程序
• 安装过程无报错，但导入TensorFlow时出现TensorRT库加载警告，且TensorFlow无法调用GPU

导入TensorFlow时的警告信息 #

2023-02-12 14:49:58.544771: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvrtc.so.11.0: cannot open shared object file: No such file or directory
2023-02-12 14:49:58.544845: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory
2023-02-12 14:49:58.544874: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.

已尝试的操作 #

• 通过命令sudo find / -name libnvinfer.so.7 2> /dev/null定位到库文件位于/usr/lib/x86_64-linux-gnu/
• 将该目录添加到LD_LIBRARY_PATH，但问题未解决

系统环境信息 #

nvidia-smi输出 #

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 515.65.01    Driver Version: 516.94       CUDA Version: 11.7     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  NVIDIA GeForce ...  On   | 00000000:01:00.0 Off |                  N/A |
| N/A   43C    P0    22W /  N/A |      0MiB /  6144MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

nvcc --version输出 #

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Fri_Jan__6_16:45:21_PST_2023
Cuda compilation tools, release 12.0, V12.0.140
Build cuda_12.0.r12.0/compiler.32267302_0

解决方案 #

1. 匹配CUDA版本与TensorFlow版本 #

TensorFlow 2.11.0官方要求的CUDA版本为11.2，当前环境nvcc显示CUDA 12.0、nvidia-smi显示CUDA 11.7，版本不匹配是GPU无法调用的核心原因。需卸载现有CUDA 12.0，安装CUDA 11.2：

• 卸载CUDA 12.0：

  sudo apt-get --purge remove cuda*
  sudo apt-get autoremove
  sudo apt-get autoclean
  

• 安装CUDA 11.2：

  wget https://developer.download.nvidia.com/compute/cuda/11.2.0/local_installers/cuda_11.2.0_460.27.04_linux.run
  sudo sh cuda_11.2.0_460.27.04_linux.run --override
  

  安装时取消勾选Driver选项（WSL2依赖Windows端驱动），仅安装Toolkit。

2. 修复TensorRT库依赖 #

警告中缺失的libnvrtc.so.11.0属于CUDA 11.x组件，安装CUDA 11.2后会自动补齐。同时安装与CUDA 11.2兼容的TensorRT 7.x版本：

sudo apt-get install libnvinfer7 libnvinfer-plugin7

• 配置环境变量，将以下内容添加到~/.bashrc：

  export LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu:/usr/local/cuda-11.2/lib64:$LD_LIBRARY_PATH
  export PATH=/usr/local/cuda-11.2/bin:$PATH
  

  执行source ~/.bashrc使配置生效。

3. 验证GPU可用性 #

重启WSL2实例后，在Python中执行以下代码验证：

import tensorflow as tf
print(tf.test.is_gpu_available())
print(tf.config.list_physical_devices('GPU'))

若输出True及GPU设备信息，说明问题已解决。

内容的提问来源于stack exchange，提问作者BARIS KURT