Activity for ParaFEM
-
Lee Margetts committed [r2281]
Minor modifications to xx21.f90
-
Minor updates to program xx16.f90
-
MINOR Change: Removed binary executable xx21
-
MAJOR Change: Added new development programs xx20 and xx21.
-
MINOR Change: Removed tool "convert_endian" from default build.
-
Added build file for Knights Landing on ARCHER.
-
Llion Evans modified a wiki page
Structure
-
Structure
-
Structure
-
Home
-
Added LFA to make-parafem
-
Added /src/tools/postprocessing/LFA as a tool to calculate effective thermal conductivity from laser flash analysis simulations.
-
Anton Shterenlikht committed [r2273]
adjusted the model size for mumu jobs on Cray to just
-
added another miniapp for Cray with netcdf + mumu
-
Minor update to /dev/xx12_b.f90
-
Minor bugfix to /dev/xx12_b/xx12_b.f90
-
Added patch to gather_scatter.f90 provided by Nicolas Gruel that seems to fix OpenMPI issue.
-
Minor change to /mpi/input.f90 SUBROUTINE READ_VARMATERIALVALUE now requires a material names in the .vmat file.
-
Bugfix to dev/xx12_te
-
Minor changes to /dev/xx12_b/ tidying of comments etc.
-
Update to /dev/xx12_te/ Now reads in .vmat files for temperature dependent material properties (as xx12_b).
-
MAJOR UPDATE to /dev/xx12_b/xx12_b.f90 Now reads in loads which can be set to vary with temperature.
-
Update to /modules/mpi/input.f90 New subroutines READ_HEADER, READ_INT, READ_REAL to read in generic data files.
-
Minor update to /dev/xx12_b/xx12_b.f90 Tidied and commented section that calculates temperature dependent material properties.
-
MAJOR CHANGE to /dev/xx12_b/xx12_b.f90 - Now reads .vmat to apply temperature dependent material properties. Should still be automatically backwards compatible with .mat files (looks for .vmat but reverts to .mat if cannot be found).
-
Update to /modules/mpi/input.f90 New subroutine READ_NTEMP to read number of temperature dependent material properties per material per property. Also updated READ_VARMATERIALVALUE to correctly read variable material properties and their associated temperatures into arrays.
-
Clarified comment
-
Update to /dev/xx12_b/ Now reads material property files that vary with temperature. Not yet applied, will currently use value for highest temperature.
-
Update to /modules/mpi/input.f90 - New subroutines READ_NMATS_NVALS & READ_NTIME to read header info from .mat and .time files. New subroutine READ_VARMATERIALVALUE to read .vmat which contains material properties that vary with temperature.
-
Update to /dev/xx12_te/ Now reads in zero stress temperature instead of assuming at 0degC.
-
Update to /dev/xx12_te/ Now outputs to ENSI format, bug present when running with MPI.
-
Update to /dev/xx12_te/ to use ENSI.NDTTR to calculate thermally induces stress etc, also now reads ENSI.GEO and ENSI.MATID as input files.
-
Update to dev/xx12_te/ to use temperatures in ensi format as well as 'Temperature.txt'. [Hardcoded switch to use different version]
-
Bugfix to /programs/dev/xx12_b/makefile
-
Updated xx12_te to read ENSI NDTTR files (but doesn't yet use this data), added SUBROUTINE READ_ENSI_SCALAR_PN to /modules/mpi/input.f90 to achieve this.
-
Created copy of rfem/rfemsolve_te/ to dev/xx12_te/ to develop to work directly with xx12.
-
Added build/sgi-mdc.inc to for make-parafem to compile on the SGI-MDC at UKAEA.
-
Flag required when compiling using ifort/12.0.1.107 and openmpi/2.0.2
-
Bug fix in /dev/xx12_b/xx12_b.f90 where counter for amp of load used j_loc instead of j_glob
-
Minor modifications to program/dev/xx1
-
Minor modifications to driver program/dev/xx1 to use new subroutine READ_ENSI_NDBND_BIN.
-
Commit of new binary I/O subroutine READ_ENSI_NDBND_BIN which is used to read the restraints in a binary format compatible with Ensight Gold. The subroutine has been tested with input deck p121_tiny.
-
Minor modifications to module src/modules/mpi/input.f90
-
Minor modifications to Ensight Gold binary readers in input.f90 and xx1.f90.
-
Fixed the binary Ensight Gold reader for the geometry (nodal coordinates and element steering array).
-
Added missing arguments to call to subroutine R...
-
committed this change as r2240.
-
fixed xx14std for OpenCoarrays/GCC.
-
I had to comment out MPI_INIT, as it's called by caf behind the scenes: svn diff xx14std.f90 Index: xx14std.f90 =================================================================== --- xx14std.f90 (revision 2239) +++ xx14std.f90 (working copy) @@ -123,20 +123,20 @@ ! intent( out ): ! numpe - integer, process number (rank) ! npes - integer, total number of processes (size) -CALL find_pe_procs( numpe, npes ) +!CALL find_pe_procs( numpe, npes ) ! CAnnot use MPI_INIT in a coarray program with ifort 16!...
-
I now have a linux system with OpenCoarrays built on top of GCC 7.0.1. So far I can't reproduce the error. Have you followed these steps: Update ParaFEM to r2239, CGPACK to r394. Build cgpack with caf, i.e. make -f Makefile-bc3-oca. Make sure to start with a clean environment, i.e. make clean first. Install cgpack, i.e. make -f Makefile-bc3-oca install. Go to parafem/parafem dir and do ./make-parafem MACHINE=caf > & build.log. You should get to: Building XX14 RELEASE make xx14std.x xx14rot_std.x...
-
Added a makefile for OpenCoarrays/GCC compiler.
-
The error reported by Sam Hewitt (UoM). It relates to dev directory xx14, which linkes ParaFEM with CGPACK (http://cgpack.sf.net).
-
xx14 build failure with GCC7
-
updated on UoB BC3
-
Mark Filipiak committed [r2237]
Add ParaFEM-PETSc interface User Guide to repos...
-
Tidied up xx19.f90 modifications.
-
Latest version of xx19.f90
-
Added example input file for tools/preprocessin...
-
Deleted executable xx19 in directory programs/d...
-
Added xx19 folder and contents to programs/src/dev
-
Added xx19 to make-parafem and xc30.inc
-
Program p129 was not working due to a change in...
-
updated xx14
-
Build xx18 xx17 xx15 only. Correct xx15_vm_non...
-
Merge from trunk.
-
Added changes in xx15_quadric_linear_hardening ...
-
Rename p_setup to p_create, which is more meani...
-
Include xx17 in build.bash messages.
-
Update README_TAGS to include xx17. Add the xx...
-
In parafem_petsc: updated comments, use VecGet...
-
Added comments.
-
Print maxdiff, maxloads, and maxdiff/maxloads o...
-
Back to the normal version of xx17.
-
Test version of xx17 to trace the diefferences ...
-
formupvw zeroes the INTENT(out) matrix to ensur...
-
Write only the final velocity ENSI file - the o...
-
(Once again) the local matrix was not being zer...
-
Comment out all changes to src/modules/mpi/gath...
-
Tidy output from xx17.
-
Correct timing in xx17.
-
PETSc added to xx17 (steady-state Navier-Stokes...
-
Use new arguments for p_print_info in xx15.
-
Added information how to use PETSc directly in ...
-
Added option to force updates in processor orde...
-
Build only xx17 in the xx group.
-
Remove the logs also when cleaning.
-
Remove original version of xx18 - this is no lo...
-
Replace xx17.f90 with xx17.F90, to use C prepro...
-
Merge from trunk.
-
Include xx17 in the xx group in make-parafem.
-
Working towards merging the petsc branch with t...
-
Corrected another makefile, changing &s to &&s.
-
The libraries are included in the LDFLAGS set i...
-
Only add the PETSc include and library flags fo...
-
The makefile for the modules had some &s instea...
-
Correct some comments.
-
Added tools/preprocessing/gaussianlds which is ...
-
Merge from trunk.
-
Make normdiff in the build.
-
Improved the program to compare (xx15 only) out...
1 >