I want to build pytorch from source with MVAPICH, and I am sure the MVAPICH is installed correctly. I meet the problem when I run the test script,
`RuntimeError: the MPI backend is not available; try to recompile the THD package with MPI support at /home/*/Downloads/pytorch/torch/lib/THD/process_group/General.cpp:17`RuntimeError: the MPI backend is not available; try to recompile the THD package with MPI support at /home/*/Downloads/pytorch/torch/lib/THD/process_group/General.cpp:17
test script
import torch.distributed as dist
dist.init_process_group(init_method='env://', backend='mpi')
I follow the setting in this pages https://discuss.pytorch.org/t/segfault-using-cuda-with-openmpi/11140, which I can build pytorch with openmpi correctly. However, I cannot use MVAPICH as backend.
Any help will be appreciated! Thank you!