diff options
Implement AA pattern for OpenCL target
Works well but function naming is getting kind of clunky, e.g. "velocity_momenta_boundary_tick_cells"
This could be hidden to a degree by proving branching wrappers for
the odd and even time step implementations. However this would not
vectorize when targeting Intel via OpenCL.
Diffstat (limited to 'boltzgen/kernel/template/collide_and_stream.cpp.mako')
0 files changed, 0 insertions, 0 deletions