Obtain the force on bounce-back boundaries

See merge request walberla/walberla!163

Michael Kuron authored 5 years ago and Christoph Rettinger committed 5 years ago

Added functionality to LBM boundaries to optionally obtain the hydrodynamic force on those boundaries. Also added tests.

undef internal macros in header files

See merge request walberla/walberla!183

Removing boost/array and boost/integer

See merge request walberla/walberla!181

moved generic blockforest creation functions from pe/CreateWorld to blockforest/Initialization.h

Removed unused LBM test case

Closes #83

See merge request walberla/walberla!176

Resolve "possible division by zero in DynamicDiffusive load balancer"

Closes #41

See merge request walberla/walberla!178

use the general classes instead. they support levelwise load
balancing via a constructor parameter

default is levelwise for backward compatibility

Resolve "introduce contact based pe load balancing"

Closes #82

See merge request walberla/walberla!174

Resolve "possible segfault in DynamicDiffusive load balancer"

Closes #42

See merge request walberla/walberla!175

Removed boost exception

See merge request walberla/walberla!177

This mainly includes an implementation for boost::uint_t::least.

Note: this could also be done with std::shared_ptr but would require
a custom deleter or std::default_delete for array types
(see also: https://stackoverflow.com/a/13062069).

Also, std::unique_ptr supports operator[] for array types.

Removed boost::array dependency in mpi::BufferDataTypeExtensions (was already partly done as part on #59)

Fix unintialized warning in Config.h when compiling with undef. sanitizer

Closes #39

See merge request walberla/walberla!170

Fix error when writing fluid field VTK output with OpenMPI version > 2.x.x

See merge request walberla/walberla!173

Added module description for pecoupling

See merge request walberla/walberla!172

- proper escaping of braces when passing JSON to python
- pass information if walberla was built with CUDA to pystencils
  to support apps that use GPU optionally

- allows for nested execution flow (i.e. solver loop inside time loop)
- timing tree integration
- parallel CUDA stream support