[mvapich-discuss] Does there exist a compile time flag or runtime function for CUDA support?

[Fang, Leo]

Hello,


We recently added support of __cuda_array_interface__<http://numba.pydata.org/numba-doc/latest/cuda/cuda_array_interface.html> to mpi4py so that CUDA-aware MPI can be seamlessly utilized. I wonder if MVAPICH provides any compile time marco or runtime function for checking whether CUDA support is enabled or not. I know that Open MPI provides such functionalities: https://www.open-mpi.org/faq/?category=runcuda#mpi-cuda-aware-support, but not sure if I missed it in the MVAPICH doc.

Thanks.


Sincerely,
Leo

---
Yao-Lung Leo Fang
Assistant Computational Scientist
Computational Science Initiative
Brookhaven National Laboratory
Bldg. 725, Room 2-169
P.O. Box 5000, Upton, NY 11973-5000
Office: (631) 344-3265
Email: leofang at bnl.gov<mailto:leofang at bnl.gov>
Website: https://leofang.github.io/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20190717/51040d19/attachment.html>