João Victor Tozatti Risso authored 7 years ago and Martin Bauer committed 7 years ago

Concept: Generate code involving the (un)packing of fields (from)to linear
(1D) arrays, i.e. (de)serialization of the field values for buffered
communication.

A linear index is generated for the buffer, by inferring the strides and
variables of the loops over fields in the AST. In the CPU, this information is
obtained through the makeLoopOverDomain function, in
pystencils/transformations/transformations.py. On CUDA, the strides of
the fields (excluding buffers) are combined with the indexing variables to infer
the indexing of the buffer.

What is supported:
    - code generation for both CPU and GPU
    - (un)packing of fields with all the memory layouts supported by
    pystencils
    - (un)packing slices of fields (from)into the buffer
    - (un)packing subsets of cell values from the fields (from)into the buffer

Limitations:

- assumes that only one buffer and one field are being operated within
each kernel, however multiple equations involving the...

- single function to create kernel for specified target
- data type creation from string - reuse numpy functionality
- bugfixes in dot display

- renaming because of clashes with types.py from other packages

- support for kernels that run over walberla ghost layers

- old method produced unused variables
- old method was not deterministic in the output code i.e. ordering of
  the introduced constants
- moveConstantsBeforeLoops transformation was also not deterministic

- pystencils can create now a non-compilable kernel that can be
  analyzed by kerncraft

- abstraction layer for selecting CUDA block and grid sizes
  - line based (was implemented before)
  - block based (new, more flexible)
-  new conditional (if/else) ast node, which is necessary for indexing schemes (guarding if)

- bugfix for CUDA kernels with variable field sizes
- extended tests for pystencils gpu kernels

problem in moveConstantBeforeLoops transformation:

--> a should end up here
{
for() {
  const int a = 5;
}
for() {
  const int a = 5
}
}

the "a" of the lower loop was not moved up, since it could not move across first loop (which is wrong)

- added sphinx files for documentation generation
- collected kernel creation functions in new "cpu" and "cudagpu" modules