How to deal with 'undefined reference to symbol 'dlclose@@GLIBC_2.2.5''?

[jidle123]

Help!! How to deal this.

Issue description

When I pass the installation and get into the test, some errors occur.

Error example

[17/19] Linking CXX executable test/ProcessGroupGlooAsyncTest
FAILED: test/ProcessGroupGlooAsyncTest 
: && /usr/bin/c++  -I"/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/include"   -I"/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/include/TH" -I"/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/include/THC"   -I"/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/include/THS" -I"/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/include/THCS"   -I"/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/include/THNN" -I"/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/include/THCUNN" -DOMPI_SKIP_MPICXX=1 -fexceptions  -std=c++11 -O3 -DNDEBUG  -L"/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/lib"   -rdynamic test/CMakeFiles/ProcessGroupGlooAsyncTest.dir/ProcessGroupGlooAsyncTest.cpp.o  -o test/ProcessGroupGlooAsyncTest  -Wl,-rpath,/home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/lib:/usr/local/cuda/lib64  -lpthread  libc10d.a  test/libc10d_cuda_test.a  libc10d.a  /home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/lib/libcaffe2_gpu.so  /home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/lib/libcaffe2.so  /home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/lib/libprotobuf.a  -lpthread  /usr/local/cuda/lib64/libcufft.so  /usr/local/cuda/lib64/libcurand.so  /usr/local/cuda/lib64/libcudnn.so.8  /usr/local/cuda/lib64/libculibos.a  /usr/local/cuda-11.1/targets/x86_64-linux/lib/libcublas_static.a  /home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/lib/libgloo_cuda.a  /home/gpu/Desktop/gpgpu-sim_distribution-master/pytorch-gpgpu-sim-master/torch/lib/tmp_install/lib/libgloo.a  /usr/local/cuda/lib64/libcudart.so && :
/usr/bin/ld: /usr/local/cuda/lib64/libculibos.a(cuos_common_posix.o): undefined reference to symbol 'dlclose@@GLIBC_2.2.5'
/usr/bin/ld: /lib/x86_64-linux-gnu/libdl.so.2: error adding symbols: DSO missing from command line
collect2: error: ld returned 1 exit status

My config: CUDA 11.1, gcc 5.3.1, python 3.8
I've looked up a few ways to fix this, which tell me to add "-ldl" in CMakeLists.txt. But nothing happened when I did this.
Has anyone else encountered this kind of problem? I would appreciate it if you could fix it.