Commit Graph

230 Commits

Author SHA1 Message Date
Martin Diehl 64a6476c1a Abaqus is complaining 2017-03-05 22:47:28 +01:00
Martin Diehl 71e2c3f203 switching to PETSc 3.7.x 2017-02-14 22:33:27 +01:00
Philip Eisenlohr d403f5f346 give feedback on invalid MSC version and path input 2017-02-10 18:33:30 -05:00
Martin Diehl 3e7ab76582 adding functionality from previous build system 2017-02-03 23:23:55 +01:00
Franz Roters 676ecb2ade correctly determine DAMASK version 2017-01-06 16:14:35 +01:00
Franz Roters 6d0e3f4f32 corrected opneMP options
fixed N/A string in version handling
2017-01-02 17:31:38 +01:00
Martin Diehl e532228b15 fixed handling of version string 2016-12-23 14:21:17 +01:00
Philip Eisenlohr 20c5ccc112 corrected (my standard) mistake in README 2016-11-29 10:02:07 -05:00
Philip Eisenlohr 94c934c8bd added patch to switch to fwbw_derivative 2016-11-29 09:52:14 -05:00
Philip Eisenlohr d6baa71ee4 added temporary patch for PETSc 3.7.x compatibility
apply with "patch -p1 < installation/patch/petsc3.7"
2016-11-24 09:51:21 -05:00
Franz Roters cbe2fb8d57 removed support for old Marc versions 2016-11-07 11:56:39 +01:00
Franz Roters b27c9388e6 adopted linker string for earlier Marc versions to work with Intel Fortran 17 2016-11-07 11:54:27 +01:00
Franz Roters 2fa4e48cf0 modified mkl libraries as to also work with Intel Fortran 17 2016-11-02 13:53:12 +01:00
Franz Roters b34a60d2f4 added support for Marc2016 2016-10-28 09:16:20 +02:00
Philip Eisenlohr eb9f6c939c condensed output, DAMASK_marc symlinks are relative, add symlink pruning 2016-08-25 12:08:19 -04:00
Martin Diehl cd63a92733 using python 2.7 has shebang will also work on mac without symlink
unless someone uses the 6 year old python 2.6, this should be save
2016-07-18 19:35:35 +02:00
Martin Diehl e7a174abc1 cleaner code 2016-07-18 19:19:51 +02:00
Martin Diehl 2a7f772d08 adopted to new installation config file 2016-07-15 11:05:27 +02:00
Martin Diehl 38bf357e7c removed obsolote information on lapack and (always outdated) version information 2016-07-14 16:18:27 +02:00
Martin Diehl 3314e5f051 updated symlinking, removed unused "rootRelation" 2016-07-14 15:11:49 +02:00
Martin Diehl 97d6ed57e1 also not needed any more 2016-06-29 20:13:00 +02:00
Martin Diehl 4b860fb083 added c routines to core module compilation 2016-05-22 17:18:15 +02:00
Martin Diehl 7ea5ce73ae line too long 2016-05-11 11:46:20 +02:00
Martin Diehl d3ee2c05b9 =removed remaining IDs 2016-05-11 11:43:21 +02:00
Martin Diehl a21e8cd863 usind explicitly python2 to run in environments with python 3 as default 2016-05-11 11:01:39 +02:00
Test User 054d9710b5 Merge branch 'development'
Conflicts:
	VERSION
2016-04-29 08:03:07 +02:00
Philip Eisenlohr c4c24d3a70 shortened comment line 2016-04-24 14:08:05 -05:00
Philip Eisenlohr 9abc0dbeeb added *.sh as possible targets for symlinking 2016-04-24 13:36:23 -05:00
Franz Roters 6bd17b3e4e integrated new version handling 2016-02-03 09:52:11 +01:00
Martin Diehl 81a7fcbe6e escape magic for string macro 2016-02-02 23:33:21 +01:00
Martin Diehl 23f1344031 for new versioning 2016-02-02 19:55:55 +01:00
Martin Diehl e298d2ebaa new version handling was broken 2016-01-28 08:56:12 +00:00
Martin Diehl e59bc9780d adopted file name change 2016-01-25 20:44:22 +00:00
Franz Roters ad71d03d13 now we are (hopefully) complete 2015-12-08 14:27:58 +00:00
Franz Roters eff2c883b4 added support for Marc 2015 and made it default 2015-12-08 14:26:27 +00:00
Franz Roters cef87db699 renamed original file 2015-12-03 07:47:28 +00:00
Franz Roters b676ed0085 support for Marc 2010 was dropped a long time ago 2015-10-20 13:32:48 +00:00
Franz Roters a087eedd54 added support for Marc2014.2 (so far installed on maws14 only) 2015-10-20 13:25:59 +00:00
Martin Diehl 7d996cfa11 fixed new IO_stringPos for Abaqus 2015-09-05 16:26:55 +00:00
Martin Diehl be4068661b fortran 2008 standard implies assume realloc-lhs 2015-09-02 13:01:39 +00:00
Martin Diehl 7a3a67601f realloc lhs was not always working, fixed now and added test for new string pos function 2015-08-31 16:30:04 +00:00
Martin Diehl eb3173c6cc ugly hack for Abaqus in prec, it seems that it links against old (10.1) ifort where IEEE_arithmetic is not avaialbe.
Took the chance and cleaned the env file and used #ifdef# statement in the wrappers because in fixed format Fortran some lines of code might be interpretated as comments.
This should not happen, but it happens for Abaqus exp
2015-04-15 18:10:56 +00:00
Franz Roters 3f55518c31 corrected spelling error in comment 2015-01-15 08:42:47 +00:00
Franz Roters a7a6ee6cff removed non OpenMP calls from DAMASK run menu as 2013.1 enforces use of OpenMP 2014-10-30 07:33:24 +00:00
Franz Roters 63b00e4fd2 use MSC_ROOT as default installation dir but allow to modify it
make Marc2014 the default version
2014-10-22 07:45:52 +00:00
Franz Roters 6cf9907cc9 added support for Marc2014 2014-10-22 07:43:06 +00:00
Pratheek Shanthraj 9b4c1fc43d added DAMASK_FEM.exe to bin_link list 2014-09-10 08:09:35 +00:00
Martin Diehl d426be49d0 fixed linker error for some ifort/f2py combinations 2014-05-21 10:07:10 +00:00
Martin Diehl 89c896678b added printing of cli arguments to seeds_fromRandom, fixed regression in compile_CoreModule and DAMASK_spectral_doxygen 2014-05-15 09:58:31 +00:00
Martin Diehl 142a72335e small changes on the deploy me / documentation scripts 2014-05-14 10:07:10 +00:00
Martin Diehl 6df66c6349 updated Marc run_damask scripts, linking against BLAS is done by Marc anyway, removed. renamed extension of original files from org to more explicit original 2014-05-07 14:29:53 +00:00
Martin Diehl 5dedad1392 fixed typo in last commit 2014-04-30 05:05:41 +00:00
Martin Diehl 6bce04df48 changed linker command slightly to ensure dynamic linking and link now agains non-parallel (but threadsafe) versions of LAPACK
(compile_CoreModule.py, Makefile, abaqus_v6.env, mod_MarcMentat)

openmp is not working for Abaqus exp, probably because the ifort 14 is not supported.
Test is now using a version without openmp
2014-04-29 16:43:59 +00:00
Martin Diehl 2346a0841e improved symbolic linking and replace deprecated print syntax 2014-04-29 14:43:04 +00:00
Philip Eisenlohr 3f5d92b9a4 polishing and shortening 2014-04-10 19:25:23 +00:00
Martin Diehl 6550729b49 now symlinking only to know extensions (cause trouble for temp files like *.py~ and the like) 2014-04-09 10:40:10 +00:00
Martin Diehl 342ca0be47 tested new scripts to update shebang, all files got same shebang (and for python files encoding) 2014-04-01 18:41:14 +00:00
Martin Diehl 188002f16e more LF, Id, and executable status changes 2014-04-01 16:58:54 +00:00
Martin Diehl 5c3e2f49d7 simplified symlinking, using python list dir method, no need to update this script anymore 2014-03-26 13:11:12 +00:00
Philip Eisenlohr 943349fdbb bunch of new scripts:
gwyddion_filter tries to smoothen out Gwyddion data sets.
vtk_fromGwyddion produces vtk mesh from Gwyddion data set.
geom_addPrimitive allows to add box, cylinder, or ellipsoidal blobs to geom file.
2014-03-04 03:34:34 +00:00
Martin Diehl f146359c04 added --prefix option to configure script to specify symlink location 2014-02-28 07:47:11 +00:00
Martin Diehl 7245e38327 like for IMKL, we probably should link ACML also against the omp version 2014-02-14 20:14:44 +00:00
Franz Roters 8ec38e8476 corrected linker options for blas 2014-02-14 15:59:25 +00:00
Martin Diehl 368de2f565 small fix on installation with imkl, need to try it out 2014-02-11 17:05:20 +00:00
Martin Diehl 25ee49ea78 fixed Abaqus test to run with new installation scheme 2014-02-05 06:38:31 +00:00
Martin Diehl 986591682c remove equals sign in rpath definition to work with mac, changed linker options in compile_CoreModule to work with linux again, please let me know if mac doesn't work anymore (it should) 2014-01-30 13:07:44 +00:00
Tias Maiti eeca9ed349 solved error: Undefined symbols for architecture x86_64: "_PyArg_ParseTupleAndKeywords" by adding arguments to LDFLAGS 2014-01-29 20:06:57 +00:00
Tias Maiti 34663daa99 rpath=argument syntax unfortunately not working on OS X, hence changed back to rpath,argument 2014-01-29 19:52:31 +00:00
Martin Diehl f82e5ec26e some small changes on the installation scripts, IMKL now works with gfortran, more verbosity and more sanity checks 2014-01-29 12:20:56 +00:00
Martin Diehl ee99216edc polishing compile_CoreModule, bug fix for Makefile 2014-01-28 07:34:13 +00:00
Martin Diehl 4648da1f32 small changes, tests still not running 2014-01-27 22:06:39 +00:00
Martin Diehl b11c64d310 polishing 2014-01-27 16:24:07 +00:00
Martin Diehl 861dc76700 fixed test for compilation of core module 2014-01-27 16:07:52 +00:00
Martin Diehl 36812cd6ea fixed some testing scripts to run with the new installation 2014-01-27 15:09:38 +00:00
Martin Diehl 02441bef0f introduced rpath to tell executable where libraries are located 2014-01-27 14:10:49 +00:00
Martin Diehl 656e3cc6ca some more small changes, including bug fix in configure + relative linking in Makefile for spectral solver 2014-01-27 10:12:29 +00:00
Martin Diehl 081d4da768 fixed path 2014-01-24 13:47:46 +00:00
Martin Diehl 554a6ab738 fixed typos and missing include 2014-01-24 13:17:54 +00:00
Martin Diehl a6dac1eceb fixed lib to use new config file 2014-01-24 13:15:01 +00:00
Martin Diehl d66b41c686 fixed installation, will test now 2014-01-24 12:30:27 +00:00
Philip Eisenlohr c9c6588f7c renamed seeds file related scripts in a (hopefully) consistent manner 2014-01-23 20:48:33 +00:00
Philip Eisenlohr 3e697e006a added (again) Intel OpenMP library to linker flag list (-liomp5). Why was that dropped..? 2014-01-22 23:57:08 +00:00
Philip Eisenlohr 380f6e1ec5 fixed syntax bugs 2014-01-22 16:03:27 +00:00
Martin Diehl e81b798662 corrected shell extension for apply_DAMASK_modifications.sh (was py) 2014-01-18 18:00:28 +00:00
Philip Eisenlohr fc7fd80e9c following convention of ".py" extension for Python scripts (now also compatible with new Makefile) 2014-01-16 21:53:14 +00:00
Martin Diehl 278f854e41 introduced functionality to get output results by point from the HDF5 file 2014-01-15 22:17:32 +00:00
Martin Diehl cf318f4392 fixed bug in lapack 2014-01-14 22:50:57 +00:00
Martin Diehl 449c66c70f did some testing... 2014-01-14 19:32:55 +00:00
Martin Diehl 01c8c434f7 finished work on the new installation scheme (if no bugs are found ;)) 2014-01-14 19:03:41 +00:00
Martin Diehl 3d69d9b927 moved DAMASK_NUM_THREADS to options, DAMASK_env.sh gets it from there 2014-01-13 10:56:54 +00:00
Martin Diehl aab0208c65 more modifications, only configure needs modifications 2014-01-12 17:29:00 +00:00
Martin Diehl 80da1c529e some more work on the consistent installation scheme, fixed bug in Makefile 2014-01-11 19:13:02 +00:00
Martin Diehl 4271f65d05 textfile containing options (like pathinfo, but extended and will be directly used by the Makefile) 2014-01-11 17:16:51 +00:00
Martin Diehl 2262098ed4 more work on new installation 2014-01-11 14:28:07 +00:00
Martin Diehl e025ac5952 fixed DAMASK_env.sh to work with linux and Mac, started to reorganize installation to be consistent for all solvers 2014-01-11 12:46:30 +00:00
Christoph Kords 9a2ad45f7a probably(?) fixed bug with Windows: comparison string should be 'Windows' not 'Windos' , I suppose 2014-01-02 15:08:09 +00:00
Martin Diehl 65de581f72 new script setup_env.py to modify DAMASK_env files, will replace setup_shellrc.py 2013-12-21 11:57:02 +00:00
Franz Roters 543639bb8f added support for Marc/Mentat2013.1 2013-12-19 13:37:35 +00:00
Martin Diehl bb0b35e466 made compilation_info.f90 windows compatible and added abaqus_v6_windows.env as a template for using DAMASK with windows.
Deleted old fftw and ACML installation instruction (can be found on damask.mpie.de)
2013-10-10 09:38:31 +00:00
Philip Eisenlohr 1593daea97 exchanged order of asking for BLASDIR and BLASTYPE. now type comes first (otherwise you may not know your options...) 2013-09-23 19:27:34 +00:00