search

hypre-space / hypre

Parallel solvers for sparse linear systems featuring multigrid methods.
https://www.llnl.gov/casc/hypre/
Other
651 stars 182 forks source link

Asynchronous device to device hipMemcpy #1079

Open tomstitt opened 4 months ago

tomstitt commented 4 months ago

Switch hipMemcpy to hipMemcpyAsync for device to device copies to avoid unneeded host-side syncs; matches cuda's performant behavior.

See https://github.com/mfem/mfem/pull/2780