Error: libxxxxxxx: cannot open shared object file: No such file or directory

出现Error: libxxxxxxx: cannot open shared object file: No such file or directory 的可能有效解决方法如下：

注：前提是本地能够找到libxxxxxxx，但一直提示：

复制代码

Could not load library libxxxxxxx. Error: libxxxxxxx: cannot open shared object file: No such file or directory
Please make sure libxxxxxxx is in your library path!

这里以libcudnn_cnn_infer.so.8为例。

先运行：

复制代码

ldconfig -p | grep cuda

得到输出：

复制代码

	libcudart.so.11.0 (libc6,x86-64) => /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudart.so.11.0
	libcudart.so.10.2 (libc6,x86-64) => /lib/libcudart.so.10.2
	libcudart.so (libc6,x86-64) => /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudart.so
	libcudadebugger.so.1 (libc6,x86-64) => /lib/x86_64-linux-gnu/libcudadebugger.so.1
	libcuda.so.1 (libc6,x86-64) => /lib/x86_64-linux-gnu/libcuda.so.1
	libcuda.so.1 (libc6) => /lib/i386-linux-gnu/libcuda.so.1

打比方说，libcudart.so.11.0，是放到/usr/local/cuda-11.3/targets/x86_64-linux/lib/目录下面的。

然后找到本地的libcudnn_cnn_infer.so.8：

复制代码

sudo find / -name "libcudnn_cnn_infer.so.8"

得到结果：

复制代码

/home/ai/miniconda3/lib/python3.7/site-packages/nvidia/cudnn/lib/libcudnn_cnn_infer.so.8
/home/ai/miniconda3/pkgs/pytorch-2.0.1-py3.10_cuda11.7_cudnn8.5.0_0/lib/python3.10/site-packages/torch/lib/libcudnn_cnn_infer.so.8
/home/ai/miniconda3/pkgs/pytorch-1.12.1-py3.7_cuda11.3_cudnn8.3.2_0/lib/python3.7/site-packages/torch/lib/libcudnn_cnn_infer.so.8
/home/ai/miniconda3/pkgs/pytorch-2.0.0-py3.10_cuda11.8_cudnn8.7.0_0/lib/python3.10/site-packages/torch/lib/libcudnn_cnn_infer.so.8

然后将找到的libcudnn_cnn_infer.so.8复制到上面通过ldconfig得到的路径：

复制代码

sudo cp -rf /home/ai/miniconda3/lib/python3.7/site-packages/nvidia/cudnn/lib/libcudnn_cnn_infer.so.8 /usr/local/cuda-11.3/targets/x86_64-linux/lib/

然后：

复制代码

sudo ldconfig

再通过ldconfig -p | grep cuda查看最新的输出：

复制代码

	libcudnn_cnn_infer.so.8 (libc6,x86-64) => /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudnn_cnn_infer.so.8
	libcudart.so.11.0 (libc6,x86-64) => /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudart.so.11.0
	libcudart.so.10.2 (libc6,x86-64) => /lib/libcudart.so.10.2
	libcudart.so (libc6,x86-64) => /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudart.so
	libcudadebugger.so.1 (libc6,x86-64) => /lib/x86_64-linux-gnu/libcudadebugger.so.1
	libcuda.so.1 (libc6,x86-64) => /lib/x86_64-linux-gnu/libcuda.so.1
	libcuda.so.1 (libc6) => /lib/i386-linux-gnu/libcuda.so.1

就能看到 libcudnn_cnn_infer.so.8了。