-
- Downloads
Added native portals support
Based on portals4 using ptl_put (RDMA) three kernels semi-, uni- and bi-directional where added. Debug options were cleaned up HAVE_XX was unified for CUDA
parent
3d7dac74
No related branches found
No related tags found
Showing
- .gitignore 11 additions, 0 deletions.gitignore
- benchmark/.gitignore 0 additions, 10 deletionsbenchmark/.gitignore
- benchmark/Makefile 104 additions, 82 deletionsbenchmark/Makefile
- benchmark/benchmark.cc 31 additions, 55 deletionsbenchmark/benchmark.cc
- benchmark/benchmark.h 4 additions, 2 deletionsbenchmark/benchmark.h
- benchmark/cmdline.cc 9 additions, 9 deletionsbenchmark/cmdline.cc
- benchmark/error.cc 16 additions, 4 deletionsbenchmark/error.cc
- benchmark/error.h 13 additions, 1 deletionbenchmark/error.h
- benchmark/gpu_nvidia.h 2 additions, 2 deletionsbenchmark/gpu_nvidia.h
- benchmark/linktest.cc 7 additions, 11 deletionsbenchmark/linktest.cc
- benchmark/memory.cc 5 additions, 5 deletionsbenchmark/memory.cc
- benchmark/memory.h 4 additions, 4 deletionsbenchmark/memory.h
- benchmark/memory_multi.cc 5 additions, 5 deletionsbenchmark/memory_multi.cc
- benchmark/output_sion.cc 4 additions, 1 deletionbenchmark/output_sion.cc
- benchmark/portals4_macros.h 18 additions, 0 deletionsbenchmark/portals4_macros.h
- benchmark/vcluster.cc 84 additions, 70 deletionsbenchmark/vcluster.cc
- benchmark/vcluster.h 13 additions, 20 deletionsbenchmark/vcluster.h
- benchmark/vcluster_cuda.cc 1 addition, 1 deletionbenchmark/vcluster_cuda.cc
- benchmark/vcluster_helper.cc 6 additions, 1 deletionbenchmark/vcluster_helper.cc
- benchmark/vcluster_mpi.cc 6 additions, 1 deletionbenchmark/vcluster_mpi.cc
Loading
Please register or sign in to comment