saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count.
introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
added some OMP FLUSH statements were necessary
replaced openmp do by forall construct where possible; this is much safer and perhaps even as fast for small loops
changed order of arrays in nearest neighbor search to make it fortran fast
constitutive.f90 and homogenization.f90 write state size out during initialization
setup/setup_processing.py is using byterecl to be compatible with binary files written out by solver
added compiler switches for gfortran and ifort to check for standard conformity
old gnu compilers <4.4 are not longer supported because they don't provide the c binding for fftw
dislocation velocities are stored in the state, so we actually now have three "parts" of the state, the basic states that are updated by "constitutive_dotState" come first, then the dependent states that are calculated by "constitutive_microstructure" follow, and finally we have a last part reserved for other variables that just use the memory reserved by the state array and are updated somewhere else.
constitutive:
LpAndItsTangent does not need the full state, but only the local state, so changed that at least for the nonlocal constitutive law
* Also added some more openmp directives to increase percentage of parallelized code.
* "implicit none" was missing in two subroutines of homogenization and constitutive.
0 : only version infos and all from "hypela2"/"umat"
1 : basic outputs from "CPFEM.f90", basic output from initialization routines, debug_info
2 : extensive outputs from "CPFEM.f90", extensive output from initialization routines
3 : basic outputs from "homogenization.f90"
4 : extensive outputs from "homogenization.f90"
5 : basic outputs from "crystallite.f90"
6 : extensive outputs from "crystallite.f90"
7 : basic outputs from the constitutive files
8 : extensive outputs from the constitutive files
If verbosity is equal to zero, all counters in debug are not set during calculation (e.g. debug_StressLoopDistribution or debug_cumDotStateTicks). This might speed up parallel calculation, because all these need critical statements which extremely slow down parallel computation.
In order to keep it like that, please follow these simple rules:
DON'T use implicit array subscripts:
example: real, dimension(3,3) :: A,B
A(:,2) = B(:,1) <--- DON'T USE
A(1:3,2) = B(1:3,1) <--- BETTER USE
In many cases the use of explicit array subscripts is inevitable for parallelization. Additionally, it is an easy means to prevent memory leaks.
Enclose all write statements with the following:
!$OMP CRITICAL (write2out)
<your write statement>
!$OMP END CRITICAL (write2out)
Whenever you change something in the code and are not sure if it affects parallelization and leads to nonconforming behavior, please ask me and/or Franz to check this.
* removed input variables in constitutive_collectDotState and constitutive_postResults that are not needed anymore (because of recent changes in constitutive_nonlocal)
Martin Diehl