Modified the gitlab-ci for benchmarking purposes
Created a new Dockerfile for benchmarking purposes, which includes the
intel compiler.
Docker file now has likwid installed as well.
Created a machine file for the benchmarking machine.
Kerncraft updated and fixed.

TODOs:
    - cleanup of the yaml files
    - Test the Skylake machine file and kerncraft_coupling for correct
      behaviour

-> one place to switch between cpu and gpu

label -> target
ceIdx -> superscript

- merged finite difference functions into one function
- put derivative operators from lbmpy into pystencils

- in-kernel Neumann boundaries
- flag-interface for boundary handling makes one flag field multiple
  boundary handlings possible
- generator: support for bitwise logical operators

- splitted existing LBM boundary handling into two parts:
    -> generic part , that is used for FD as well and moved it to pystencils
    -> LBM specific part - remained in lbmpy

- bugfixes

- step class for LB phasefield generic enough to work with 3-phase and
  N-phase models
- cahn hilliard can either be solved by LBM or by finite differences
- 3 phase model can be solved with rho phase or without

- now an lbMethod or updateRule can be passed in directly
- made method/update rule/ast function independent i.e. no function
  uses the same parameters as on of the others

- 3 phase model works now with step sytem

- plotting
- calculation of force on boundary
- vtk outpu

- previously all objects where cached by id()
- for waLBerla simulations in each time step a new np.array view
  is created from the waLBerla field. Each of these views has a
  different id -> caching did not work for waLBerla setups
- changed hash for numpy arrays: instead of id, a tuple of
  (dataPtr, strides, shapes) is used as hash input

- Step working for serial CPU scenarios

- scaling interface width eta instead of surface tensions tau to correct
  interface profile & surface tensions

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...

Fixed demo_heat_equation notebook.

Fixed toDot.

Remove unnecessary comments.

because boundary handling does its own flag management