Commit Graph

27 Commits

Author SHA1 Message Date
Martin Diehl 90440b50b7 prec provides IEEE module 2019-05-11 11:19:06 +02:00
Martin Diehl 72110d509c one implicit none is enough 2019-05-09 08:30:03 +02:00
Martin Diehl 1ff55b6374 not supported by ancient compiler needed for Abaqus 2019-03-26 21:44:25 +01:00
Martin Diehl 81b971374b no need to use pInt 2019-03-07 19:34:27 +01:00
Martin Diehl d17c1a6d17 comment was wrong 2019-03-07 18:37:22 +01:00
Martin Diehl 7a083c4098 [skip ci] more general names and procedure 2019-03-07 11:02:27 +01:00
Martin Diehl a965c46025 improved functions for floating point comparison
- less stric tolerance for comparison to zero
- better readable
- avoid "merge" on optional arguments (not safe)
- "equal" and "notEqual" are now symmetric (assignment of <= and < is
arbitrary)
2019-03-06 15:57:42 +01:00
Martin Diehl 51f8b1961f simplify integer precision handling
essentially, it should be ok to always use the default integer (which is
32 bit unless using MSC.Marc) and use 64 bit integer only for special cases where an overflow
could happen
2019-03-06 15:52:52 +01:00
Martin Diehl 977f61452b compiler-independent defintion of real and integer kinds
real(8) does not neccessarily mean a real with 8 byte (but for gfortran
and ifort it does)
2019-03-06 15:25:47 +01:00
Martin Diehl 17a682e883 P/epsijk parameter has nothing to do with precision 2019-02-01 08:53:57 +01:00
Martin Diehl 40d38ebf55 added rotation conversions
modified versions from 3Drotations code (available on GitHub) by Marc De Graef
2018-12-08 08:02:55 +01:00
Martin Diehl 18db5f5652 not needed 2018-09-18 05:20:08 +02:00
Franz Roters 38fd517c44 Merge remote-tracking branch 'remotes/origin/36-faster-file-handling-for-material-config-use-stream-access-instead-of-line-wise-reading' into development 2018-08-27 15:13:16 +02:00
Martin Diehl 271b9ba76b intersting note ... 2018-08-22 23:28:47 +02:00
Martin Diehl 52002f654e to converge at one point to one (or two) string lenth values 2018-08-22 09:14:16 +02:00
Martin Diehl 3e4c878304 using shared interface for spectral and FEM solver
group_scalar seems inappropriate as integers are also scalars. renamed
to group_float (is actually usually of double precision).

think about better name, types should have a t prefix. tgroupFloat?
2018-08-20 16:09:40 +02:00
Philip Eisenlohr ba215ed9ea small polish 2018-08-13 18:27:51 -04:00
Philip Eisenlohr 8edeeaf213 renaming of p_vec and friends to names based on "group" 2018-08-03 19:39:14 -04:00
Martin Diehl 31311d63db gfortran 8 was complaining, asigned pointer is not contiguous 2018-05-16 00:09:33 +02:00
Martin Diehl d80e15bd76 compiler_options and compiler_version supported by new Intel compiler 2018-02-02 12:36:09 +01:00
Martin Diehl 09a66d918d (in)equality comparison for double was far too tolerant 2017-11-21 09:24:25 +01:00
Martin Diehl e9440ccd6f suppress warning when compiling with gfortran 2017-10-05 16:35:34 +02:00
Philip Eisenlohr f5816a643d added offsetDeltaState to type-definition and respect this when applying deltaState 2017-09-29 17:44:10 -04:00
Martin Diehl b35ff67f99 using IEEE_is_NAN and IEEE_quiet_NaN instead of hand-written solution, will not work for gfortran < 5 2017-05-04 00:32:44 +02:00
Martin Diehl d86074829c Merge branch 'development' into cmake 2017-02-03 19:28:23 +01:00
Chen Zhang 672c8d0606 Merge remote branch 'origin/development' into cmake 2016-04-18 09:11:29 -04:00
Chen Zhang e33594cb44 change code folder to src 2016-03-04 14:43:29 -05:00