CarpetX’s OpenMP parallelization scheme tiles components, and calls scheduled functions once per tile. The new cGH fields tile_min and tile_max describe the local tile.

See https://bitbucket.org/cactuscode/cactus/pull-requests/104/cactus-new-cgh-fields-tile_min-tile_max

--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2446/cactus-new-cgh-fields-tile_min-tile_max