223031012
/
DAMASK_EICMD
mirror of https://github.com/achalhp/DAMASK_EICMD.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
DAMASK_EICMD/code/prec.f90

113 lines
6.1 KiB
Fortran
Raw Normal View History

changed comments in prec and IO to be read by doxygen, fine tuned output of doxygen and added precompilation for documentation
2012-08-09 16:31:53 +05:30
!--------------------------------------------------------------------------------------------------
! $Id$
!--------------------------------------------------------------------------------------------------
some comments and minor improvements. removed tol_gravityNodePos from prec because it's not used tol_math_check is now used by spectral driver to check rotation
2013-09-14 16:29:35 +05:30
!> @author Franz Roters, Max-Planck-Institut für Eisenforschung GmbH
!> @author Philip Eisenlohr, Max-Planck-Institut für Eisenforschung GmbH
!> @author Christoph Kords, Max-Planck-Institut für Eisenforschung GmbH
!> @author Martin Diehl, Max-Planck-Institut für Eisenforschung GmbH
some changes related to new state fixed assigning of wrong output in J2 model
2014-05-08 20:25:19 +05:30
!> @author Luv Sharma, Max-Planck-Institut für Eisenforschung GmbH
some comments and minor improvements. removed tol_gravityNodePos from prec because it's not used tol_math_check is now used by spectral driver to check rotation
2013-09-14 16:29:35 +05:30
!> @brief setting precision for real and int type depending on makros "FLOAT" and "INT"
!> @details setting precision for real and int type and for DAMASK_NaN. Definition is made
!! depending on makros "FLOAT" and "INT" defined during compilation
changed comments in prec and IO to be read by doxygen, fine tuned output of doxygen and added precompilation for documentation
2012-08-09 16:31:53 +05:30
!--------------------------------------------------------------------------------------------------
utf8 encoding output to screen now for all solvers using intel ifort>10 or any gfortran (old versions might not support it, but we don't use them anyway)
2012-04-28 16:16:41 +05:30
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
module prec
changed comments in prec and IO to be read by doxygen, fine tuned output of doxygen and added precompilation for documentation
2012-08-09 16:31:53 +05:30
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
implicit none
private
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
#if (FLOAT==4)
fixed floating point exception in nonlocal (calculation of dUpper at several places)
2014-06-27 01:31:38 +05:30
#if defined(Spectral) || defined(FEM)
SPECTRAL SOLVER AND OWN FEM DO NOT SUPPORT SINGLE PRECISION, STOPPING COMPILATION
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
#endif
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
integer, parameter, public :: pReal = 4 !< floating point single precition (was selected_real_kind(6,37), number with 6 significant digits, up to 1e+-37)
again some changes regarding creation of NaN, ieee-conforming way is not in the complilers yet.
2012-10-02 22:23:03 +05:30
#ifdef __INTEL_COMPILER
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
real(pReal), parameter, public :: DAMASK_NaN = Z'7F800001' !< quiet NaN for single precision (from http://www.hpc.unimelb.edu.au/doc/f90lrm/dfum_035.html, copy can be found in documentation/Code/Fortran)
again some changes regarding creation of NaN, ieee-conforming way is not in the complilers yet.
2012-10-02 22:23:03 +05:30
#endif
#ifdef __GFORTRAN__
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
real(pReal), parameter, public :: DAMASK_NaN = real(Z'7F800001', pReal) !< quiet NaN for single precision (from http://www.hpc.unimelb.edu.au/doc/f90lrm/dfum_035.html, copy can be found in documentation/Code/Fortran)
#endif
#elif (FLOAT==8)
fixed bug causing trouble with ifort 11 (abaqus)
2012-08-29 00:40:54 +05:30
integer, parameter, public :: pReal = 8 !< floating point double precision (was selected_real_kind(15,300), number with 15 significant digits, up to 1e+-300)
again some changes regarding creation of NaN, ieee-conforming way is not in the complilers yet.
2012-10-02 22:23:03 +05:30
#ifdef __INTEL_COMPILER
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
real(pReal), parameter, public :: DAMASK_NaN = Z'7FF8000000000000' !< quiet NaN for double precision (from http://www.hpc.unimelb.edu.au/doc/f90lrm/dfum_035.html, copy can be found in documentation/Code/Fortran)
again some changes regarding creation of NaN, ieee-conforming way is not in the complilers yet.
2012-10-02 22:23:03 +05:30
#endif
#ifdef __GFORTRAN__
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
real(pReal), parameter, public :: DAMASK_NaN = real(Z'7FF8000000000000', pReal) !< quiet NaN for double precision (from http://www.hpc.unimelb.edu.au/doc/f90lrm/dfum_035.html, copy can be found in documentation/Code/Fortran)
#endif
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
#else
made error in case of invalid precision more verbose
2013-09-18 19:29:42 +05:30
NO SUITABLE PRECISION FOR REAL SELECTED, STOPPING COMPILATION
small polishing, exchanged stops in math.f90 with calls to IO_error
2012-02-13 19:38:07 +05:30
#endif
let's see whether this is the thing to do..!?!?
2007-03-20 19:25:22 +05:30
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
#if (INT==4)
integer, parameter, public :: pInt = 4 !< integer representation 32 bit (was selected_int_kind(9), number with at least up to +- 1e9)
#elif (INT==8)
integer, parameter, public :: pInt = 8 !< integer representation 64 bit (was selected_int_kind(12), number with at least up to +- 1e12)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
#else
made error in case of invalid precision more verbose
2013-09-18 19:29:42 +05:30
NO SUITABLE PRECISION FOR INTEGER SELECTED, STOPPING COMPILATION
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
#endif
integer, parameter, public :: pLongInt = 8 !< integer representation 64 bit (was selected_int_kind(12), number with at least up to +- 1e12)
some comments and minor improvements. removed tol_gravityNodePos from prec because it's not used tol_math_check is now used by spectral driver to check rotation
2013-09-14 16:29:35 +05:30
real(pReal), parameter, public :: tol_math_check = 1.0e-8_pReal !< tolerance for internal math self-checks (rotation)
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
some comments and minor improvements. removed tol_gravityNodePos from prec because it's not used tol_math_check is now used by spectral driver to check rotation
2013-09-14 16:29:35 +05:30
type, public :: p_vec !< variable length datatype used for storage of state
changed pointer attribute to allocatable in p_vec and p_intvec
2014-06-25 21:38:49 +05:30
real(pReal), dimension(:), allocatable :: p
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
end type p_vec
added version information to all files do NOT edit text like this: $Id: constitutive_phenopowerlaw.f90 406 2009-08-31 14:13:10Z MPIE\f.roters $
2009-08-31 20:39:15 +05:30
added int vec pointer for more efficient memory allocation of homog and micro counting (elemsOfHomogMicro) in material.f90
2013-02-25 18:43:52 +05:30
type, public :: p_intvec
changed pointer attribute to allocatable in p_vec and p_intvec
2014-06-25 21:38:49 +05:30
integer(pInt), dimension(:), allocatable :: p
added int vec pointer for more efficient memory allocation of homog and micro counting (elemsOfHomogMicro) in material.f90
2013-02-25 18:43:52 +05:30
end type p_intvec
introduced new state structure, compile with STATE=NEW
2014-04-15 14:50:38 +05:30
#ifdef NEWSTATE
!http://stackoverflow.com/questions/3948210/can-i-have-a-pointer-to-an-item-in-an-allocatable-array
type, public :: tState
changed pointer attribute to allocatable in tState. since the pointers are only used for allocation, allocatables should do the same job more efficiently
2014-06-25 19:52:08 +05:30
integer(pInt) :: sizeState,sizeDotState
logical :: nonlocal
real(pReal), allocatable, dimension(:) :: atolState
real(pReal), allocatable, dimension(:,:) :: state, & ! material points, state size
dotState, &
state0, &
partionedState0, &
subState0, &
state_backup, &
deltaState, &
previousDotState, &
previousDotState2, &
dotState_backup, &
RK4dotState
real(pReal), allocatable, dimension(:,:,:) :: RKCK45dotState
introduced new state structure, compile with STATE=NEW
2014-04-15 14:50:38 +05:30
end type
#endif
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
public :: &
prec_init
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
contains
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
changed comments in prec and IO to be read by doxygen, fine tuned output of doxygen and added precompilation for documentation
2012-08-09 16:31:53 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief reporting precision and checking if DAMASK_NaN is set correctly
!--------------------------------------------------------------------------------------------------
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
subroutine prec_init
changed comments in prec and IO to be read by doxygen, fine tuned output of doxygen and added precompilation for documentation
2012-08-09 16:31:53 +05:30
use, intrinsic :: iso_fortran_env ! to get compiler_version and compiler_options (at least for gfortran 4.6 at the moment)
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
implicit none
updated doxygen config file for FEM solvers, replaced some external statements, removed debugging options from abaqus_v6.env causing Abaqus to crash (didn't detect the bug so far), added new homogenization test to automator, added more debug options + comments to Makefile
2013-02-13 23:24:56 +05:30
external :: &
quit
added doxygen documentation to material.f90 and marked read-only quantities as protected where possible, removed substituted "call flush" by "flush"
2013-01-18 17:00:52 +05:30
write(6,'(/,a)') ' <<<+- prec init -+>>>'
write(6,'(a)') ' $Id$'
made code standard conform to Fortran 2008 (ignoring warning concerning comments beyond character 132). Basically, changing "x" format specifier to "Nx" ("1x") plus removing $ format specifier 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
2012-02-01 00:48:55 +05:30
#include "compilation_info.f90"
introduced new state structure, compile with STATE=NEW
2014-04-15 14:50:38 +05:30
#ifdef NEWSTATE
write(6,'(a)') 'Using new state structure'
#endif
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
write(6,'(a,i3)') ' Bytes for pReal: ',pReal
write(6,'(a,i3)') ' Bytes for pInt: ',pInt
write(6,'(a,i3)') ' Bytes for pLongInt: ',pLongInt
corrected NaN in prec_single.f90 (was already silent but with typo), polished output and added switch in order to check range on all files except of prec.f90 (in which NaN is defined)
2012-05-11 18:16:17 +05:30
write(6,'(a,e10.3)') ' NaN: ', DAMASK_NaN
corrected PETSc macro expansion
2013-01-10 21:06:55 +05:30
write(6,'(a,l3,/)') ' NaN /= NaN: ',DAMASK_NaN/=DAMASK_NaN
introduced new state structure, compile with STATE=NEW
2014-04-15 14:50:38 +05:30
moved option for UTF-8 encoding to spectral interface as this is now called before prec init. if UTF-8 encoding is needed for FEM solver (old compilers might not support it), add the same statement to the corresponding interface routine
2012-10-04 19:52:39 +05:30
if (DAMASK_NaN == DAMASK_NaN) call quit(9000)
added version information to all files do NOT edit text like this: $Id: constitutive_phenopowerlaw.f90 406 2009-08-31 14:13:10Z MPIE\f.roters $
2009-08-31 20:39:15 +05:30
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
end subroutine prec_init
added version information to all files do NOT edit text like this: $Id: constitutive_phenopowerlaw.f90 406 2009-08-31 14:13:10Z MPIE\f.roters $
2009-08-31 20:39:15 +05:30
added error for to IO to substitute stop statement in kdtree2 explicitly defined all functions in as either public or private in the modules to have a quick overview on all functions and parameters that are available
2012-03-06 20:22:48 +05:30
end module prec