|
| subroutine | gs_mpi::gs_mpi_init (this, send_pe, recv_pe) |
| | Initialise MPI based communication method See gs_comm.f90 for details.
|
| |
| subroutine | gs_mpi::gs_mpi_init_vec (this) |
| | Allocate the fused vector exchange buffers, sized for GS_VEC_NC components. Deferred to the first fused exchange, see gs_comm_t.
|
| |
| subroutine | gs_mpi::gs_mpi_free (this) |
| | Deallocate MPI based communication method.
|
| |
| subroutine | gs_mpi::gs_nbsend_mpi (this, u, n, tag, deps, strm) |
| | Post non-blocking send operations.
|
| |
| subroutine | gs_mpi::gs_nbrecv_mpi (this, tag) |
| | Post non-blocking receive operations.
|
| |
| subroutine | gs_mpi::gs_nbwait_mpi (this, u, n, op, strm) |
| | Wait for non-blocking operations.
|
| |
| subroutine | gs_mpi::gs_nbsend_vec_mpi (this, u, n, nc, tag, deps, strm) |
| | Post non-blocking sends for a fused nc-component exchange.
|
| |
| subroutine | gs_mpi::gs_nbrecv_vec_mpi (this, tag, nc) |
| | Post non-blocking receives for a fused nc-component exchange.
|
| |
| subroutine | gs_mpi::gs_nbwait_vec_mpi (this, u, n, nc, op, strm) |
| | Wait for a fused nc-component exchange and reduce each received slab.
|
| |