Assign the positions in state vectors, where to fetch from or save to
Solver specific part to identify the actual data locations
to send use within the communication buffers.
Todo
this should be pattern specific, with typed exchange we
do not need to store the pos, as it is done inside MPI
with the defined datatype
Access to state array: SAVE
Why? Because we need to work on the pdfs on which the
compute kernel was working. The kernel stored to SAVE
All actions taking place on the state array after the kernel
need to be performed with SAVE access
Nodes of different colours represent the following:
Solid arrows point from a procedure to one which it calls. Dashed
arrows point from an interface to procedures which implement that interface.
This could include the module procedures in a generic interface or the
implementation in a submodule of an interface in a parent module.
Nodes of different colours represent the following:
Solid arrows point from a procedure to one which it calls. Dashed
arrows point from an interface to procedures which implement that interface.
This could include the module procedures in a generic interface or the
implementation in a submodule of an interface in a parent module.