82dabe29c1
mesh: separate kind for DAMASK and PETSc integers
2022-01-13 16:30:42 +01:00
29530da579
use correct kind of constants for calls to MPI/PETSc
2022-01-13 13:50:30 +01:00
a7417a7ad7
default integer, PETSc integer, and MPI integer might be different
2022-01-13 12:02:33 +01:00
3fb5bd459c
pInt leftovers
2022-01-13 07:44:15 +01:00
43ae4983ed
fixed misaligned/too short "writing..." statements
2021-11-15 12:58:59 -05:00
69843d0833
Merge branch 'development' into pretty-print-init
2021-11-15 12:39:14 -05:00
da9fdf53d2
consistent indentation and line-spacings in reporting
2021-11-15 12:35:44 -05:00
8138d3a180
style adjustments
2021-11-11 23:59:48 +01:00
ba31ff7263
do not rely on ML per default
2021-11-11 23:23:14 +01:00
b4ff89f304
improved test
2021-10-27 22:54:26 +02:00
a00c6743c3
symbolic notation in numerics.yaml
...
- p_i: integration order
- p_s: shape function order
ensure working combination (p_s = p_i: full integration, p_s = p_i+1:
reduced integration)
2021-10-26 11:51:08 +02:00
5ac592eb9e
initialize L_(i/p), L_(i,p)0
2021-07-22 12:31:10 +02:00
69cc0b528b
do calculation for initialization
2021-07-22 11:38:03 +02:00
5b66db8a39
only mechanics at the moment
...
will be extended, but most likely differently
2021-07-22 10:12:43 +02:00
eb834b635d
unify notation with grid
...
consistent style: Symbols, not descriptions.
also removed untested/unused loginc functionality. Once load cases are
written in YAML, we can introduce the scaling as in DAMASK_grid
2021-07-22 08:48:44 +02:00
173a5f8e55
less public/unneeded variables
2021-07-22 07:48:01 +02:00
de7ef43095
documenting/silencing gfortran
2021-07-21 23:49:24 +02:00
58a7921966
no need for temp, pack instead of reshape for 1D array
2021-07-21 22:45:04 +02:00
b817c620a3
let the computer count
2021-07-21 21:25:33 +02:00
23077fd58c
testing coordinates
...
I don't understand why this pattern exists (with w(2) = 3, w(3) = 2),
but it exists
2021-07-21 19:42:53 +02:00
5a04a1d661
let the computer do the counting
2021-07-21 19:41:14 +02:00
c56979e2ad
single source of truth
2021-07-21 18:47:19 +02:00
000007ba59
better test automatically
2021-07-21 18:06:37 +02:00
b1ba64e6af
understanding floating point precision
...
double can use 16 or 17 significant figures
- if the leading number is high (above 5 or so), only 16 figures are
significant.
- a trailing 1 is never significant
2021-07-21 18:04:50 +02:00
f25dee4c3a
quad points updated and rounded based on literature values
2021-07-21 14:34:18 +02:00
a919ac0aea
Merge branch 'write_ipdisplacements' of git.damask.mpie.de:damask/DAMASK into write_ipdisplacements
2021-07-20 15:35:48 +02:00
86d918367f
Round quadrature points
2021-07-20 15:34:13 +02:00
4a2e62fdb2
Merge branch 'development' into 'write_ipdisplacements'
...
# Conflicts:
# src/mesh/mesh_mech_FEM.f90
2021-07-19 17:19:40 +00:00
a654cd4fb1
To avoid fortran double variable truncation for quadrature points
2021-07-19 17:07:43 +02:00
97203ff551
document and keep definitions together
2021-07-18 23:30:30 +02:00
f9edeb40a5
descriptive names
2021-07-17 11:50:21 +02:00
ed6b1be352
solver handles terminally ill
2021-07-16 20:43:08 +02:00
3f0eafd640
first step towards separating of mechanics, thermal, and damage
2021-07-16 17:53:11 +02:00
2f1fa8292b
unify style to majority of occurences
2021-07-16 10:30:45 +02:00
3830492791
Merge remote-tracking branch 'origin/development' into write_ipdisplacements
2021-07-15 17:34:44 +02:00
4d6a9a51ed
correct logic
2021-07-15 17:32:41 +02:00
94843becda
writing ip displacements (algorithm contain bugs)
2021-07-14 19:33:54 +02:00
136a4b1377
PETSc defines are rather complicated
...
now mpi_f08 can be used on newer PETSc installations if old MPI modules
are not exposed
2021-07-09 18:48:25 +02:00
637f78bd52
old name (for PETSc < 3.15)
2021-07-09 14:50:29 +02:00
139f2c177a
use MPI_f08 if possible
...
most PETSc installations provide outdated MPI (f90 version)
MPI_COMM_WORLD is now of derived type (Fortran 08 style)
PETSC_COMM_WORLD is the plain integer (f90 style) alias.
Note that HDF5 is assumed to have f90 interfaces
2021-07-08 16:27:37 +02:00
a0ffbc5f17
little polishing
...
- same names as for grid
- 'local' naming misleading here (probably for x_local etc also not a
good choice)
- following names in PETSc notation (https://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/DMPLEX/DMPlexGetPointLocal.html )
- using allocation with bounds to simplify access
- NOTE: The nodal coordinates are not what damask.Result expects except
for simple, 1 IP elements!
2021-07-02 23:49:06 +02:00
b322ed3e03
feauture to write nodal displacements to hdf5 file added
2021-06-25 20:46:21 +02:00
36bf68b336
Out of plane coordinates for 2D elements are updated with dummy values
2021-06-17 20:26:37 +02:00
b9d4217d86
permutationStar111 for 4th order integration is updataed
2021-06-16 14:58:26 +02:00
e10701b32a
easier to understand
2021-06-14 15:21:10 +02:00
5f01b076d2
This change makes the test work for GNU too
...
Probably something was uninitialized and GNU was accessing unallocated memory
2021-06-14 14:58:14 +02:00
c64f9eb805
hardcoding '3' is an issue for 2d simulations
2021-06-11 20:42:31 +02:00
6cd0d79c61
old merge not reflected
2021-06-02 14:08:12 +02:00
91015fd07c
write out undeformed configuration
2021-05-31 19:41:22 +02:00
650fbd7509
Initial nodal and ip coordinates added to the HDF5 output file
2021-05-31 15:01:42 +02:00
Martin Diehl