223031012
/
DAMASK_EICMD
mirror of https://github.com/achalhp/DAMASK_EICMD.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

better readable

This commit is contained in:
Martin Diehl 2019-06-08 10:38:10 +02:00
parent bc8ced72a4
commit aeb57b2fb6
1 changed files with 36 additions and 34 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

64
src/mesh_grid.f90
View File

@ -13,11 +13,11 @@ module mesh
use DAMASK_interface use DAMASK_interface
use IO use IO
use debug use debug
use numerics
use discretization use discretization
use geometry_plastic_nonlocal use geometry_plastic_nonlocal
use FEsolving use FEsolving
implicit none implicit none
private private
@ -43,45 +43,48 @@ contains
!-------------------------------------------------------------------------------------------------- !--------------------------------------------------------------------------------------------------
!> @brief initializes the mesh by calling all necessary private routines the mesh module !> @brief reads the geometry file to obtain information on discretization
!! Order and routines strongly depend on type of solver
!-------------------------------------------------------------------------------------------------- !--------------------------------------------------------------------------------------------------
subroutine mesh_init(ip,el) subroutine mesh_init(ip,el)
integer, intent(in), optional :: el, ip integer, intent(in), optional :: el, ip ! for compatibility reasons
include 'fftw3-mpi.f03' include 'fftw3-mpi.f03'
integer(C_INTPTR_T) :: devNull, local_K, local_K_offset real(pReal), dimension(3) :: &
integer :: ierr, worldsize, j mySize !< domain size of this process
real(pReal), dimension(:,:), allocatable :: IPvolume integer, dimension(3) :: &
integer, dimension(:), allocatable :: & myGrid !< domain grid of this process
integer, dimension(:), allocatable :: &
microstructureAt, & microstructureAt, &
homogenizationAt homogenizationAt
logical :: myDebug logical :: myDebug
integer :: j
integer(C_INTPTR_T) :: &
devNull, z, z_offset
write(6,'(/,a)') ' <<<+- mesh init -+>>>' write(6,'(/,a)') ' <<<+- mesh init -+>>>'
myDebug = iand(debug_level(debug_mesh),debug_levelBasic) /= 0
myDebug = (iand(debug_level(debug_mesh),debug_levelBasic) /= 0_pInt)
call mesh_spectral_read_grid(grid,geomSize,microstructureAt,homogenizationAt) call mesh_spectral_read_grid(grid,geomSize,microstructureAt,homogenizationAt)
call MPI_comm_size(PETSC_COMM_WORLD, worldsize, ierr) if(worldsize>grid(3)) call IO_error(894, ext_msg='number of processes exceeds grid(3)')
if(ierr /=0_pInt) call IO_error(894_pInt, ext_msg='MPI_comm_size')
if(worldsize>grid(3)) call IO_error(894_pInt, ext_msg='number of processes exceeds grid(3)')
call fftw_mpi_init() call fftw_mpi_init
devNull = fftw_mpi_local_size_3d(int(grid(3),C_INTPTR_T), & devNull = fftw_mpi_local_size_3d(int(grid(3),C_INTPTR_T), &
int(grid(2),C_INTPTR_T), & int(grid(2),C_INTPTR_T), &
int(grid(1),C_INTPTR_T)/2+1, & int(grid(1),C_INTPTR_T)/2+1, &
PETSC_COMM_WORLD, & PETSC_COMM_WORLD, &
local_K, & ! domain grid size along z z, & ! domain grid size along z
local_K_offset) ! domain grid offset along z z_offset) ! domain grid offset along z
grid3 = int(local_K) grid3 = int(z)
grid3Offset = int(local_K_offset) grid3Offset = int(z_offset)
size3 = geomSize(3)*real(grid3,pReal) /real(grid(3),pReal) size3 = geomSize(3)*real(grid3,pReal) /real(grid(3),pReal)
size3Offset = geomSize(3)*real(grid3Offset,pReal)/real(grid(3),pReal) size3Offset = geomSize(3)*real(grid3Offset,pReal)/real(grid(3),pReal)
myGrid = [grid(1:2),grid3]
mySize = [geomSize(1:2),size3]
microstructureAt = microstructureAt(product(grid(1:2))*grid3Offset+1: & microstructureAt = microstructureAt(product(grid(1:2))*grid3Offset+1: &
product(grid(1:2))*(grid3Offset+grid3)) ! reallocate/shrink in case of MPI product(grid(1:2))*(grid3Offset+grid3)) ! reallocate/shrink in case of MPI
@ -89,28 +92,27 @@ subroutine mesh_init(ip,el)
product(grid(1:2))*(grid3Offset+grid3)) ! reallocate/shrink in case of MPI product(grid(1:2))*(grid3Offset+grid3)) ! reallocate/shrink in case of MPI
mesh_ipCoordinates = mesh_build_ipCoordinates([grid(1:2),grid3],[geomSize(1:2),size3],grid3Offset) mesh_ipCoordinates = mesh_build_ipCoordinates(myGrid,mySize,grid3Offset)
if (myDebug) write(6,'(a)') ' Built IP coordinates'; flush(6) if (myDebug) write(6,'(a)') ' Built IP coordinates'; flush(6)
allocate(IPvolume(1,grid(1)*grid(2)*grid3),source=product([geomSize(1:2),size3]/real([grid(1:2),grid3]))) call geometry_plastic_nonlocal_setIPvolume( &
call geometry_plastic_nonlocal_setIPvolume(IPvolume) reshape([(product(mySize/real(myGrid,pReal)),j=1,product(myGrid))],[1,product(myGrid)]))
call geometry_plastic_nonlocal_setIParea(mesh_build_ipAreas([geomSize(1:2),size3],[grid(1:2),grid3])) call geometry_plastic_nonlocal_setIParea(mesh_build_ipAreas(mySize,myGrid))
call geometry_plastic_nonlocal_setIPareaNormal(mesh_build_ipNormals(grid(1)*grid(2)*grid3)) call geometry_plastic_nonlocal_setIPareaNormal(mesh_build_ipNormals(product(myGrid)))
call geometry_plastic_nonlocal_setIPneighborhood(mesh_spectral_build_ipNeighborhood([grid(1:2),grid3])) call geometry_plastic_nonlocal_setIPneighborhood(mesh_spectral_build_ipNeighborhood(myGrid))
if (myDebug) write(6,'(a)') ' Built nonlocal geometry'; flush(6) if (myDebug) write(6,'(a)') ' Built nonlocal geometry'; flush(6)
if (debug_e < 1 .or. debug_e > grid(1)*grid(2)*grid3) & if (debug_e < 1 .or. debug_e > product(myGrid)) &
call IO_error(602,ext_msg='element') ! selected element does not exist call IO_error(602,ext_msg='element') ! selected element does not exist
if (debug_i /= 1) & if (debug_i /= 1) &
call IO_error(602,ext_msg='IP') ! selected element does not have requested IP call IO_error(602,ext_msg='IP') ! selected element does not have requested IP
FEsolving_execElem = [1,grid(1)*grid(2)*grid3] ! parallel loop bounds set to comprise all DAMASK elements FEsolving_execElem = [1,product(myGrid)] ! parallel loop bounds set to comprise all elements
allocate(FEsolving_execIP(2,grid(1)*grid(2)*grid3), source=1) ! parallel loop bounds set to comprise from first IP... allocate(FEsolving_execIP(2,product(myGrid)),source=1) ! parallel loop bounds set to comprise the only IP
forall (j = 1:grid(1)*grid(2)*grid3) FEsolving_execIP(2,j) = 1 ! ...up to own IP count for each element
call discretization_init(homogenizationAt,microstructureAt, & call discretization_init(homogenizationAt,microstructureAt, &
reshape(mesh_ipCoordinates,[3,grid(1)*grid(2)*grid3]),& reshape(mesh_ipCoordinates,[3,product(myGrid)]), &
mesh_spectral_build_nodes([grid(1:2),grid3],[geomSize(1:2),size3],grid3Offset)) mesh_spectral_build_nodes(myGrid,mySize,grid3Offset))
end subroutine mesh_init end subroutine mesh_init