Skip to content
Snippets Groups Projects
Commit 19c9b9f5 authored by Gaurav Garg's avatar Gaurav Garg
Browse files

Remove thread-MPI limitation for GPU direct PME-PP communication 

Allows use of direct-GPU communication for PP-PME communication when
running with "real" MPI, including on multiple compute nodes,
through new CUDA-aware MPI communication code paths.

Implements part of #2891
Refs: #2915 #3960
parent 2fcd9717
No related branches found
No related tags found
Hide whitespace changes
Inline Side-by-side
Showing
with 431 additions and 124 deletions
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment