Re: [SIESTA-L] phonon calculations
Hi, For example, If you have 10 atoms, you can split them into two or more different calculation by just changing MD.FCfirst and MD.FClast and keep other part same.Than combine two calculation.FC files into one calculation.FC file. first calculation MD.FCfirst 1 MD.FClast 3 second calculation MD.FCfirst 4 MD.FClast 10 On 2018-08-15 12:49, Najmeh Honari wrote: Dear siesta users, I terms of phonon calculations, I want to conduct different FC runs on smaller subsets of atoms in central unit cell, on different processors. And then combine the different FC files. Would you let me know how I can do it? Regards, Najmeh Honari
Re: [SIESTA-L] optimisation monoclinic to orthorhombic
I tried increasing k-points to 4,6, and 8, but the results is similar, monoclinic structure changed to orthorhombic one. I will try to look into pseudopotential for oxygen. Thanks, Tao On 2018-05-17 10:23, T. Liu wrote: Dear all, I made a test on CuO (monoclinic beta = 99 expt.), it accomplished but the structure changed into an orthorhombic one (beta = 90). Does it mean that the pseudopotential or/and basis set of Cu has some problems or anything else? I also tried use 3p of Cu as semicore, but it goes to beta=90 or SCF doesn't converge. Thanks. SystemName CuO SystemLabel CuO NumberOfSpecies 2 NumberOfAtoms 8 %block ChemicalSpeciesLabel 1 29 Cu 2 8 O %endblock ChemicalSpeciesLabel LatticeConstant 1.0 Ang %block LatticeVectors 4.6529998779 0.00 0.00 0.00 3.410858 0.00 -0.8413048067 0.00 5.0382405853 %endblock LatticeVectors AtomicCoordinatesFormat Fractional %block AtomicCoordinatesAndAtomicSpecies 0.25000 0.25000 0.0 1 0.75000 0.75017 0.0 1 0.75000 0.25000 0.49985 1 0.25000 0.75017 0.49985 1 0.0 0.41598 0.24992 2 0.0 0.58402 0.74977 2 0.5 0.91633 0.24992 2 0.5 0.08393 0.74977 2 %endblock AtomicCoordinatesAndAtomicSpecies MeshCutoff 800 Ry MD.VariableCell T DM.Tolerance 0.10E-04 MD.MaxForceTol 0.001 eV/Ang MD.MaxStressTol 0.01 GPa XC.Functional GGA XC.authors PBEsol MaxSCFIterations 300 # Default value SCFMustConverge T# Default value PAO.BasisSize TZP MD.TypeOfRunCG MD.NumCGsteps 200 MD.VariableCell T DM.MixingWeight 0.15 %block kgrid_Monkhorst_Pack 2 0 0 0.5 0 2 0 0.5 0 0 2 0.5 %endblock kgrid_Monkhorst_Pack Regards, Tao
[SIESTA-L] optimisation monoclinic to orthorhombic
Dear all, I made a test on CuO (monoclinic beta = 99 expt.), it accomplished but the structure changed into an orthorhombic one (beta = 90). Does it mean that the pseudopotential or/and basis set of Cu has some problems or anything else? I also tried use 3p of Cu as semicore, but it goes to beta=90 or SCF doesn't converge. Thanks. SystemName CuO SystemLabel CuO NumberOfSpecies 2 NumberOfAtoms 8 %block ChemicalSpeciesLabel 1 29 Cu 2 8 O %endblock ChemicalSpeciesLabel LatticeConstant 1.0 Ang %block LatticeVectors 4.6529998779 0.00 0.00 0.00 3.410858 0.00 -0.8413048067 0.00 5.0382405853 %endblock LatticeVectors AtomicCoordinatesFormat Fractional %block AtomicCoordinatesAndAtomicSpecies 0.25000 0.25000 0.0 1 0.75000 0.75017 0.0 1 0.75000 0.25000 0.49985 1 0.25000 0.75017 0.49985 1 0.0 0.41598 0.24992 2 0.0 0.58402 0.74977 2 0.5 0.91633 0.24992 2 0.5 0.08393 0.74977 2 %endblock AtomicCoordinatesAndAtomicSpecies MeshCutoff 800 Ry MD.VariableCell T DM.Tolerance 0.10E-04 MD.MaxForceTol 0.001 eV/Ang MD.MaxStressTol 0.01 GPa XC.Functional GGA XC.authors PBEsol MaxSCFIterations 300 # Default value SCFMustConverge T# Default value PAO.BasisSize TZP MD.TypeOfRunCG MD.NumCGsteps 200 MD.VariableCell T DM.MixingWeight 0.15 %block kgrid_Monkhorst_Pack 2 0 0 0.5 0 2 0 0.5 0 0 2 0.5 %endblock kgrid_Monkhorst_Pack Regards, Tao
[SIESTA-L] Data in SystemLabel.vectors
Dear All, After using Vibra to get the vibrational frequency, I can get such a file called SystemLabel.vectors, and there are many coordinates (or I may call them vectors) in the file as following, k= 0.000.000.00 Eigenvector = 1 Frequency=-0.108924 Eigenmode (real part) 0.6144E-02 0.2053E-01 0.3226E+00 0.6144E-02 0.2053E-01 0.3226E+00 0.6269E-02 0.2095E-01 0.3291E+00 0.6269E-02 0.2095E-01 0.3291E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.4731E-02 0.1581E-01 0.2484E+00 0.1187E-02 0.3968E-02 0.6235E-01 0.1187E-02 0.3968E-02 0.6235E-01 0.1187E-02 0.3968E-02 0.6235E-01 0.1187E-02 0.3968E-02 0.6235E-01 Eigenmode (imaginary part) 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 0.E+00 -0.E+00 Could anyone tell me what these coordinates are about please? Are they the deviations (xyz) from the equilibrium position of each atom under this normal mode? What's the Unit for these coordinates? Thanks. Regards, Tao
[SIESTA-L] FC calculation continuation
Dear All I am working on a FC calculation on a large system. I know that we can split the calculation over every atoms in the unit cell. But in my case the FC calculation stop at the 6th step because of the walltime, and I don't want to repeat the previous 5 steps with more CPU cores again (usesavedata has been used). Could anyone please tell me if it is possible that we do only one displacement on a certain direction (let say +z direction) of one atom to calculate the force constants, and then combine the FC file together with the previous 5 steps of the same atom. Thanks. Regards, Tao
Re: [SIESTA-L] compilation error for siesta trunk 572
Dear Nick, Thanks. I tried, but it gave me same error message. I think I need to try 4.1-b1, So I used the same arch file to compile 4.1-b1, but met the following error message. Could you please suggest any solution? I searched, but didn't find any useful information. mpifc -c -g -O3 `FoX/FoX-config --fcflags` /home/tl439/siesta-4.1-b1/Src/densematrix.f mpifc -c -g -O3 `FoX/FoX-config --fcflags` /home/tl439/siesta-4.1-b1/Src/files.f mpifc -c -g -O3 `FoX/FoX-config --fcflags` /home/tl439/siesta-4.1-b1/Src/get_kpoints_scale.f90 mpifc -c -g -O3 `FoX/FoX-config --fcflags` /home/tl439/siesta-4.1-b1/Src/m_gamma.F90 mpifc -c -g -O3 `FoX/FoX-config --fcflags` /home/tl439/siesta-4.1-b1/Src/m_spin.F90 /home/tl439/siesta-4.1-b1/Src/m_spin.F90:60.29: type(tSpin), public :: spin 1 Error: Object 'spin' at (1) must have the SAVE attribute for default initialization of a component make: *** [m_spin.o] Error 1 Regards, Tao On 2016-10-18 15:02, Nick Papior wrote: With intel compilers you should probably not compile with so high optimizations. First try and lower the optimization (-O2 for instance). 2016-10-18 15:28 GMT+02:00 T. Liu <tl...@cam.ac.uk>: Dear Nick, Thanks for your help. I compiled it again without "~", It works! Instead, I met an old error message which I have already posted before, but haven't got an answer. There is no DM file in the folder. the bottom of output file Setting up quadratic distribution... ExtMesh (bp) on 0 = 147 x 144 x 134 = 2836512 PhiOnMesh: Number of (b)points on node 0 = 3800 PhiOnMesh: nlist on node 0 = 985489 stepf: Fermi-Dirac step function Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 8 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 15 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 1 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 2 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 3 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 4 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 5 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 6 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 7 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 9 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 10 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 11 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 12 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 13 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 14 Stopping Program from Node: 0 INPUT file is SystemName H2O SystemLabel H2O NumberOfSpecies 2 NumberOfAtoms 24 %block ChemicalSpeciesLabel 1 1 H 2 8 O %endblock ChemicalSpeciesLabel LatticeConstant 6.3612 Ang %block LatticeVectors 1.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.0 %endblock LatticeVectors AtomicCoordinatesFormat Fractional %block AtomicCoordinatesAndAtomicSpecies 0.29160 0.29160 0.30827 1 0.95832 0.95832 0.30827 1 0.04168 0.20840 0.05835 1 0.20840 0.04168 0.05835 1 0.54168 0.54168 0.22506 1 0.70839 0.70839 0.22506 1 0.79160 0.45831 0.97498 1 0.45830 0.79160 0.97498 1 0.45830 0.29160 0.47498 1 0.79160 0.95831 0.47498 1 0.70839 0.20839 0.72506 1 0.54170 0.04168 0.72506 1 0.20840 0.54168 0.55835 1 0.04168 0.70840 0.55835 1 0.95832 0.45832 0.80827 1 0.29160 0.79160 0.80827 1 0.12504 0.12504 0.14171 2 0.62504 0.62504 0.14171 2 0.62504 0.12504 0.64171 2 0.12504 0.62504 0.64171 2 0.37496 0.37496 0.39163 2 0.87496 0.87496 0.39163 2 0.87496 0.37496 0.89163 2 0.37496 0.87496 0.89163 2 %endblock AtomicCoordinatesAndAtomicSpecies MD.TypeOfRun CG MD.NumCGsteps 200 SCFMustConverge T XC.Functional GGA XC.authors PBE %block PAO.Basis # DZPQ-8.5 O 3 n=2 0 2 E 50. 7.5 8.0 2.7 n=2 1 2 E 10. 8.3 8.5 2.6 n=3 2 1 E 40. 8.3 Q 6.8 0.22 8.5 H 2 n=1 0 2 E 50. 8.3 8.5 2.2 n=2 1 1 E 20. 7.8 Q 6.5 0.9 8.0 %endblock PAO.Basis DM.Tolerance 0.10E-04 MD.MaxForceTol 0.001 eV/Ang MD.MaxStressTol 0.01 GPa MD.VariableCell T kgrid_cutoff 10.00 Ang MeshCutoff 400 Ry I don't think there is any error in the input file, because it works for siesta 4.0. So I guess it must be still something wrong with the compilation? Regards, Tao On 2016-10-18 13:48, Nick Papior wrote: From the output of siesta, it is compiled in parallel, (it says PARALLEL). It may be that you are not using the same mpirun as that shipped with mpifc? Maybe you should be using mpiifort? Thes
Re: [SIESTA-L] compilation error for siesta trunk 572
Dear Nick, Thanks for your help. I compiled it again without "~", It works! Instead, I met an old error message which I have already posted before, but haven't got an answer. There is no DM file in the folder. the bottom of output file Setting up quadratic distribution... ExtMesh (bp) on 0 = 147 x 144 x 134 = 2836512 PhiOnMesh: Number of (b)points on node 0 = 3800 PhiOnMesh: nlist on node 0 = 985489 stepf: Fermi-Dirac step function Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:8 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 15 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:1 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:2 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:3 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:4 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:5 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:6 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:7 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node:9 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 10 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 11 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 12 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 13 Bad DM normalization: Qtot, Tr[D*S] = 64. 9.15895573 Stopping Program from Node: 14 Stopping Program from Node:0 INPUT file is SystemName H2O SystemLabel H2O NumberOfSpecies 2 NumberOfAtoms 24 %block ChemicalSpeciesLabel 11 H 28 O %endblock ChemicalSpeciesLabel LatticeConstant 6.3612 Ang %block LatticeVectors 1.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.0 %endblock LatticeVectors AtomicCoordinatesFormat Fractional %block AtomicCoordinatesAndAtomicSpecies 0.29160 0.29160 0.30827 1 0.95832 0.95832 0.30827 1 0.04168 0.20840 0.05835 1 0.20840 0.04168 0.05835 1 0.54168 0.54168 0.22506 1 0.70839 0.70839 0.22506 1 0.79160 0.45831 0.97498 1 0.45830 0.79160 0.97498 1 0.45830 0.29160 0.47498 1 0.79160 0.95831 0.47498 1 0.70839 0.20839 0.72506 1 0.54170 0.04168 0.72506 1 0.20840 0.54168 0.55835 1 0.04168 0.70840 0.55835 1 0.95832 0.45832 0.80827 1 0.29160 0.79160 0.80827 1 0.12504 0.12504 0.14171 2 0.62504 0.62504 0.14171 2 0.62504 0.12504 0.64171 2 0.12504 0.62504 0.64171 2 0.37496 0.37496 0.39163 2 0.87496 0.87496 0.39163 2 0.87496 0.37496 0.89163 2 0.37496 0.87496 0.89163 2 %endblock AtomicCoordinatesAndAtomicSpecies MD.TypeOfRunCG MD.NumCGsteps 200 SCFMustConverge T XC.Functional GGA XC.authors PBE %block PAO.Basis # DZPQ-8.5 O 3 n=2 0 2 E 50. 7.5 8.0 2.7 n=2 1 2 E 10. 8.3 8.5 2.6 n=3 2 1 E 40. 8.3 Q 6.8 0.22 8.5 H 2 n=1 0 2 E 50. 8.3 8.5 2.2 n=2 1 1 E 20. 7.8 Q 6.5 0.9 8.0 %endblock PAO.Basis DM.Tolerance 0.10E-04 MD.MaxForceTol 0.001 eV/Ang MD.MaxStressTol 0.01 GPa MD.VariableCell T kgrid_cutoff 10.00 Ang MeshCutoff 400 Ry I don't think there is any error in the input file, because it works for siesta 4.0. So I guess it must be still something wrong with the compilation? Regards, Tao On 2016-10-18 13:48, Nick Papior wrote: From the output of siesta, it is compiled in parallel, (it says PARALLEL). It may be that you are not using the same mpirun as that shipped with mpifc? Maybe you should be using mpiifort? These are long shots... but... Secondly, it may be a problem with the scheduler and the "~" in the path. Could you try and move the executable to another folder? If either of the above does not work, I think you should consult your local HPC administrator, from the output you provide SIESTA is enabled in parallel mode. 2016-10-18 14:41 GMT+02:00 T. Liu <tl...@cam.ac.uk>: Dear Nick, Thanks for your reply. it is mpirun -ppn $mpi_tasks_per_node -np $np $application $options translated as mpirun -ppn 16 -np 16 /home/tl439/~siesta-maint/siesta/trunk/Obj/siesta < h2o.relax.fdf > h2o.relax.out I also tried mpirun -np 16 /home/tl439/~siesta-maint/siesta/trunk/Obj/siesta < h2o.relax.fdf > h2o.relax.out it doesn't work n
Re: [SIESTA-L] compilation error for siesta trunk 572
Dear Nick, Thanks for your reply. it is mpirun -ppn $mpi_tasks_per_node -np $np $application $options translated as mpirun -ppn 16 -np 16 /home/tl439/~siesta-maint/siesta/trunk/Obj/siesta < h2o.relax.fdf > h2o.relax.out I also tried mpirun -np 16 /home/tl439/~siesta-maint/siesta/trunk/Obj/siesta < h2o.relax.fdf > h2o.relax.out it doesn't work neither. I used same execution command for siesta 4.0, it works. Regards, Tao On 2016-10-18 13:13, Nick Papior wrote: Please show us your execution command line. 2016-10-18 14:08 GMT+02:00 T. Liu <tl...@cam.ac.uk>: Dear Nick, I followed your suggestion, and there is no error for the compilation. But I only can run a serial mode on a parallel version. here is a piece of the top of output file *** *** * WELCOME TO SIESTA * *** *** * WELCOME TO SIESTA * *** *** * WELCOME TO SIESTA * *** * WELCOME TO SIESTA * *** reinit: Reading from standard input reinit: Reading from standard input ... ... * Running in serial mode Siesta Version: trunk-572 Architecture : x86_64-unknown-linux-gnu--unknown Compiler flags: mpifc -g -O3 PP flags : -DMPI -DFC_HAVE_FLUSH -DFC_HAVE_ABORT Libraries : -L/usr/local/Cluster-Apps/intel/mkl/11.3.3.210/compilers_and_libraries_2016.33.210/linux/mkl/lib/intel64 [1] -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_lp64 -lpthread -lm PARALLEL version Could you suggest what to do please? Regards, Tao On 2016-10-14 12:18, T. Liu wrote: Dera Nick, Thanks for your help. It works by adding .c and using latest SCALAPCK. Regards, Tao On 2016-10-14 11:40, Nick Papior wrote: Try add .c to the SUFFIXES list. 2016-10-14 12:32 GMT+02:00 T. Liu <tl...@cam.ac.uk>: Dear all, I can compile 4.0 with the following arch.make arch.make # # Copyright (C) 1996-2016 The SIESTA group # This file is distributed under the terms of the # GNU General Public License: see COPYING in the top directory # or http://www.gnu.org/copyleft/gpl.txt [2] [1]. # See Docs/Contributors.txt for a list of contributors. # .SUFFIXES: .SUFFIXES: .f .F .o .a .f90 .F90 SIESTA_ARCH=x86_64-unknown-linux-gnu--unknown FPP= FPP_OUTPUT= FC=mpifc RANLIB=ranlib SYS=nag SP_KIND=4 DP_KIND=8 KINDS=$(SP_KIND) $(DP_KIND) FFLAGS=-g -O2 FPPFLAGS= -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI LDFLAGS= ARFLAGS_EXTRA= FCFLAGS_fixed_f= FCFLAGS_free_f90= FPPFLAGS_fixed_F= FPPFLAGS_free_F90= BLAS_LIBS=-lblas LAPACK_LIBS=-llapack BLACS_LIBS= SCALAPACK_LIBS= #DUMMY_FOX=--enable-dummy NETCDF_LIBS= NETCDF_INTERFACE= COMP_LIBS=dc_lapack.a #LIBS=$(SCALAPACK_LIBS) $(BLACS_LIBS) $(LAPACK_LIBS) $(BLAS_LIBS) $(NETCDF_LIBS) LIBS=-L${MKLROOT}/lib/intel64 -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_lp64 -lpthread -lm #SIESTA needs an F90 interface to MPI #This will give you SIESTA's own implementation #If your compiler vendor offers an alternative, you may change #to it here. MPI_INTERFACE=libmpi_f90.a MPI_INCLUDE= . #Dependency rules are created by autoconf according to whether #discrete preprocessing is necessary or not. .F.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_fixed_F) $< .F90.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_free_F90) $< .f.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_fixed_f) $< .f90.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_free_f90) $< -- Then I just changed COMP_LIBS=dc_lapack.a to COMP_LIBS += libsiestaLAPACK.a to compile the Siesta trunk 572. It seems it almost finished but an error shows up at the bottom (error message shown below,"gfortran: sockets.o: No such file or directory"). Could you please suggest what happen and how to fix that? Thanks. ... ... mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/m_monitor.F90 mpifc -c -g -O2 `FoX/FoX-config --fcflags` /home/tl439/~siesta-maint/siesta/trunk/Src/nag.f mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/pxf.F90 mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/siesta.F mpifc -o siesta automatic_cell.o atom_options.o arw.o atomlwf.o bands.o basis_enthalpy.o bessph.o bonds.o born_charge.o cellxc_mod.o cgwf.o chkdim.o chkgmx.o chempot.o coceri.o coxmol.o cross.o compute_norm.o denmat.o denmatlomem.o detover.o dfscf.o m_diagon_opt.o diagon.o digcel.o fft.o dhscf.o constr.o diagk_file.o diagg.o diagk.o diagkp.o diag2g.o diag2k.o diag3g.o diag3k.o diagpol.o diagsprl.o dipole.o dismin.o dnaefs.o doping_uniform.o dot.o m_efield.o egandd.o ener3.o ener3lomem.o errorf.o extrapolon.o m_fixed.o interpolation.o gradient.o gradien
Re: [SIESTA-L] compilation error for siesta trunk 572
Dear Nick, I followed your suggestion, and there is no error for the compilation. But I only can run a serial mode on a parallel version. here is a piece of the top of output file *** *** * WELCOME TO SIESTA * *** *** * WELCOME TO SIESTA * *** *** * WELCOME TO SIESTA * *** * WELCOME TO SIESTA * *** reinit: Reading from standard input reinit: Reading from standard input ... ... * Running in serial mode Siesta Version: trunk-572 Architecture : x86_64-unknown-linux-gnu--unknown Compiler flags: mpifc -g -O3 PP flags : -DMPI -DFC_HAVE_FLUSH -DFC_HAVE_ABORT Libraries : -L/usr/local/Cluster-Apps/intel/mkl/11.3.3.210/compilers_and_libraries_2016.33.210/linux/mkl/lib/intel64 -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_lp64 -lpthread -lm PARALLEL version Could you suggest what to do please? Regards, Tao On 2016-10-14 12:18, T. Liu wrote: Dera Nick, Thanks for your help. It works by adding .c and using latest SCALAPCK. Regards, Tao On 2016-10-14 11:40, Nick Papior wrote: Try add .c to the SUFFIXES list. 2016-10-14 12:32 GMT+02:00 T. Liu <tl...@cam.ac.uk>: Dear all, I can compile 4.0 with the following arch.make arch.make # # Copyright (C) 1996-2016 The SIESTA group # This file is distributed under the terms of the # GNU General Public License: see COPYING in the top directory # or http://www.gnu.org/copyleft/gpl.txt [1]. # See Docs/Contributors.txt for a list of contributors. # .SUFFIXES: .SUFFIXES: .f .F .o .a .f90 .F90 SIESTA_ARCH=x86_64-unknown-linux-gnu--unknown FPP= FPP_OUTPUT= FC=mpifc RANLIB=ranlib SYS=nag SP_KIND=4 DP_KIND=8 KINDS=$(SP_KIND) $(DP_KIND) FFLAGS=-g -O2 FPPFLAGS= -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI LDFLAGS= ARFLAGS_EXTRA= FCFLAGS_fixed_f= FCFLAGS_free_f90= FPPFLAGS_fixed_F= FPPFLAGS_free_F90= BLAS_LIBS=-lblas LAPACK_LIBS=-llapack BLACS_LIBS= SCALAPACK_LIBS= #DUMMY_FOX=--enable-dummy NETCDF_LIBS= NETCDF_INTERFACE= COMP_LIBS=dc_lapack.a #LIBS=$(SCALAPACK_LIBS) $(BLACS_LIBS) $(LAPACK_LIBS) $(BLAS_LIBS) $(NETCDF_LIBS) LIBS=-L${MKLROOT}/lib/intel64 -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_lp64 -lpthread -lm #SIESTA needs an F90 interface to MPI #This will give you SIESTA's own implementation #If your compiler vendor offers an alternative, you may change #to it here. MPI_INTERFACE=libmpi_f90.a MPI_INCLUDE= . #Dependency rules are created by autoconf according to whether #discrete preprocessing is necessary or not. .F.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_fixed_F) $< .F90.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_free_F90) $< .f.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_fixed_f) $< .f90.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_free_f90) $< -- Then I just changed COMP_LIBS=dc_lapack.a to COMP_LIBS += libsiestaLAPACK.a to compile the Siesta trunk 572. It seems it almost finished but an error shows up at the bottom (error message shown below,"gfortran: sockets.o: No such file or directory"). Could you please suggest what happen and how to fix that? Thanks. ... ... mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/m_monitor.F90 mpifc -c -g -O2 `FoX/FoX-config --fcflags` /home/tl439/~siesta-maint/siesta/trunk/Src/nag.f mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/pxf.F90 mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/siesta.F mpifc -o siesta automatic_cell.o atom_options.o arw.o atomlwf.o bands.o basis_enthalpy.o bessph.o bonds.o born_charge.o cellxc_mod.o cgwf.o chkdim.o chkgmx.o chempot.o coceri.o coxmol.o cross.o compute_norm.o denmat.o denmatlomem.o detover.o dfscf.o m_diagon_opt.o diagon.o digcel.o fft.o dhscf.o constr.o diagk_file.o diagg.o diagk.o diagkp.o diag2g.o diag2k.o diag3g.o diag3k.o diagpol.o diagsprl.o dipole.o dismin.o dnaefs.o doping_uniform.o dot.o m_efield.o egandd.o ener3.o ener3lomem.o errorf.o extrapolon.o m_fixed.o interpolation.o gradient.o gradientlomem.o grdsam.o hsparse.o idiag.o initatom.o inver.o iodm_netcdf.o iodmhs_netcdf.o iogrid_netcdf.o iolwf.o iozm.o ipack.o iopipes.o iosockets.o iowfs_netcdf.o kgrid.o kgridinit.o kinefsm.o ksv.o ksvinit.o linpack.o local_DOS.o madelung.o
Re: [SIESTA-L] Noncollinear SpinPolarized calculation
Dear all, I met same error by siesta trunk 572, and there is no DM file. here is my .fdf SystemName H2O SystemLabel H2O NumberOfSpecies 2 NumberOfAtoms 3 %block ChemicalSpeciesLabel 11 H 28 O %endblock ChemicalSpeciesLabel LatticeConstant 1.0 Ang %block LatticeVectors 6.36120 0.0 0.0 0.0 6.36120 0.0 0.0 0.0 6.36120 %endblock LatticeVectors AtomicCoordinatesFormat Fractional %block AtomicCoordinatesAndAtomicSpecies 0.0 0.0 0.0 2 0.0 0.0 0.2 1 0.0 0.2 0.0 1 %endblock AtomicCoordinatesAndAtomicSpecies XC.Functional GGA XC.authors PBE kgrid_cutoff 10.00 Ang MeshCutoff 300 Ry Please suggest what to do? Regards, Tao On 2016-10-05 09:50, Nick Papior wrote: Please delete the old DM file. 2016-10-05 10:34 GMT+02:00 Raha khalili: Dear Dr. Papier, Many thanks for the helping and so sorry for my too many emails. I've installed siesta-4.0 on my system but the problem persists with this error: stepf: Fermi-Dirac step function Bad DM normalization: Qtot, Tr[D*S] = 18. 17.90742510 Stopping Program from Node: 0. Here are attached my input and output files. Could you help me please to solve this problem? Thank you in advance. Best regards, Raha On Tue, Sep 27, 2016 at 11:22 PM, Nick Papior wrote: Please try siesta-4.0 instead. 2016-09-27 8:57 GMT+02:00 Raha khalili : Dear Nick, Many thanks for the reply. I tried to do calculation based on the manual but it doesn't work. If I set tolerance at 0.001, I get an error. The input and output files are attached here. Could you give a look at them please? Thank you in advance, Best regards Raha On Mon, Sep 26, 2016 at 7:52 PM, Nick Papior wrote: Yes, you can use non colinear spin configuration. Please see the manual for instructions. Please note that the SpinPolarized flag refers to colinear spin configuration. -- Kind regards Nick Papior On 26 Sep 2016 17:52, "Raha khalili" wrote: Dear all, I want to do a Noncollinear SpinPolarized calculation for my structure using Siesta but I am wondering whether it is possible to do such calclulations using siesta or not. If yes, could you give me please an example or reference paper? I have also one queation about the block of DM.InitSpin. As I undrestood I can tune spin direction using two angles of phi and teta. But when I set SpinPolarized equel T and change angles, for instance for bulk Fe, the total energy and density of states are similar. Any help will be really appreciated. Best regards, Raha -- Khadijeh Khalili Nanotechnology group, Physics department, university of Mazandaran Babolsar, Iran -- Khadijeh Khalili Nanotechnology group, Physics department, university of Mazandaran Babolsar, Iran -- Kind regards Nick -- Khadijeh Khalili Nanotechnology group, Physics department, university of Mazandaran Babolsar, Iran -- Kind regards Nick
Re: [SIESTA-L] compilation error for siesta trunk 572
Dera Nick, Thanks for your help. It works by adding .c and using latest SCALAPCK. Regards, Tao On 2016-10-14 11:40, Nick Papior wrote: Try add .c to the SUFFIXES list. 2016-10-14 12:32 GMT+02:00 T. Liu <tl...@cam.ac.uk>: Dear all, I can compile 4.0 with the following arch.make arch.make # # Copyright (C) 1996-2016 The SIESTA group # This file is distributed under the terms of the # GNU General Public License: see COPYING in the top directory # or http://www.gnu.org/copyleft/gpl.txt [1]. # See Docs/Contributors.txt for a list of contributors. # .SUFFIXES: .SUFFIXES: .f .F .o .a .f90 .F90 SIESTA_ARCH=x86_64-unknown-linux-gnu--unknown FPP= FPP_OUTPUT= FC=mpifc RANLIB=ranlib SYS=nag SP_KIND=4 DP_KIND=8 KINDS=$(SP_KIND) $(DP_KIND) FFLAGS=-g -O2 FPPFLAGS= -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI LDFLAGS= ARFLAGS_EXTRA= FCFLAGS_fixed_f= FCFLAGS_free_f90= FPPFLAGS_fixed_F= FPPFLAGS_free_F90= BLAS_LIBS=-lblas LAPACK_LIBS=-llapack BLACS_LIBS= SCALAPACK_LIBS= #DUMMY_FOX=--enable-dummy NETCDF_LIBS= NETCDF_INTERFACE= COMP_LIBS=dc_lapack.a #LIBS=$(SCALAPACK_LIBS) $(BLACS_LIBS) $(LAPACK_LIBS) $(BLAS_LIBS) $(NETCDF_LIBS) LIBS=-L${MKLROOT}/lib/intel64 -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_lp64 -lpthread -lm #SIESTA needs an F90 interface to MPI #This will give you SIESTA's own implementation #If your compiler vendor offers an alternative, you may change #to it here. MPI_INTERFACE=libmpi_f90.a MPI_INCLUDE= . #Dependency rules are created by autoconf according to whether #discrete preprocessing is necessary or not. .F.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_fixed_F) $< .F90.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_free_F90) $< .f.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_fixed_f) $< .f90.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_free_f90) $< -- Then I just changed COMP_LIBS=dc_lapack.a to COMP_LIBS += libsiestaLAPACK.a to compile the Siesta trunk 572. It seems it almost finished but an error shows up at the bottom (error message shown below,"gfortran: sockets.o: No such file or directory"). Could you please suggest what happen and how to fix that? Thanks. ... ... mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/m_monitor.F90 mpifc -c -g -O2 `FoX/FoX-config --fcflags` /home/tl439/~siesta-maint/siesta/trunk/Src/nag.f mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/pxf.F90 mpifc -c -g -O2 `FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/siesta.F mpifc -o siesta automatic_cell.o atom_options.o arw.o atomlwf.o bands.o basis_enthalpy.o bessph.o bonds.o born_charge.o cellxc_mod.o cgwf.o chkdim.o chkgmx.o chempot.o coceri.o coxmol.o cross.o compute_norm.o denmat.o denmatlomem.o detover.o dfscf.o m_diagon_opt.o diagon.o digcel.o fft.o dhscf.o constr.o diagk_file.o diagg.o diagk.o diagkp.o diag2g.o diag2k.o diag3g.o diag3k.o diagpol.o diagsprl.o dipole.o dismin.o dnaefs.o doping_uniform.o dot.o m_efield.o egandd.o ener3.o ener3lomem.o errorf.o extrapolon.o m_fixed.o interpolation.o gradient.o gradientlomem.o grdsam.o hsparse.o idiag.o initatom.o inver.o iodm_netcdf.o iodmhs_netcdf.o iogrid_netcdf.o iolwf.o iozm.o ipack.o iopipes.o iosockets.o iowfs_netcdf.o kgrid.o kgridinit.o kinefsm.o ksv.o ksvinit.o linpack.o local_DOS.o madelung.o mesh.o meshphi.o meshdscf.o memory.o meshsubs.o metaforce.o m_supercell.o mulliken.o minvec.o naefs.o m_new_dm.o normalize_dm.o ordern.o outcell.o outcoor.o m_fft_gpfa.o paste.o pdos.o pdosg.o pdosk.o pdoskp.o phirphi.o pixmol.o plcharge.o projected_DOS.o propor.o m_partial_charges.o randomg.o reclat.o redcel.o reinit.o reord.o reordpsi.o rhoofd.o rhoofdsp.o rhooda.o savepsi.o shaper.o timer_tree.o timer.o vmb.o vmat.o vmatsp.o volcel.o cdiag.o rdiag.o cgvc.o cgvc_zmatrix.o m_convergence.o iocg.o ioeig.o iofa.o iokp.o iomd.o kpoint_pdos.o typecell.o ofc.o poison.o readsp.o radfft.o write_md_record.o kpoint_grid.o find_kgrid.o proximity_check.o state_init.o siesta_move.o setup_hamiltonian.o compute_dm.o mixer.o scfconvergence_test.o post_scf_work.o state_analysis.o write_subs.o siesta_init.o struct_init.o siesta_options.o read_options.o siesta_geom.o siesta_analysis.o siesta_end.o siesta_forces.o io.o sparse_matrices.o coor.o atm_transfer.o broadcast_basis.o eggbox.o dsyevds.o zheevds.o optical.o phirphi_opt.o reoptical.o transition_rate.o initparallel.o show_distribution.o setspatial.o setatomnodes.o uncell.o cart2frac.o obc.o precision.o sys.o m_cell.o files.o spatial.o parallel.o parallelsubs.o parsing.o chemical.o atom.o atmparams.o m_mpi_utils.o fdf_extra.o m_iorho.o atmfuncs.o listsc.o memoryinfo.o m_memory.o sorting.o atomlist.o atm_types.o old_atmfuncs.o radial.o alloc.o spher_harm.
[SIESTA-L] compilation error for siesta trunk 572
Dear all, I can compile 4.0 with the following arch.make arch.make # # Copyright (C) 1996-2016 The SIESTA group # This file is distributed under the terms of the # GNU General Public License: see COPYING in the top directory # or http://www.gnu.org/copyleft/gpl.txt. # See Docs/Contributors.txt for a list of contributors. # .SUFFIXES: .SUFFIXES: .f .F .o .a .f90 .F90 SIESTA_ARCH=x86_64-unknown-linux-gnu--unknown FPP= FPP_OUTPUT= FC=mpifc RANLIB=ranlib SYS=nag SP_KIND=4 DP_KIND=8 KINDS=$(SP_KIND) $(DP_KIND) FFLAGS=-g -O2 FPPFLAGS= -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI LDFLAGS= ARFLAGS_EXTRA= FCFLAGS_fixed_f= FCFLAGS_free_f90= FPPFLAGS_fixed_F= FPPFLAGS_free_F90= BLAS_LIBS=-lblas LAPACK_LIBS=-llapack BLACS_LIBS= SCALAPACK_LIBS= #DUMMY_FOX=--enable-dummy NETCDF_LIBS= NETCDF_INTERFACE= COMP_LIBS=dc_lapack.a #LIBS=$(SCALAPACK_LIBS) $(BLACS_LIBS) $(LAPACK_LIBS) $(BLAS_LIBS) $(NETCDF_LIBS) LIBS=-L${MKLROOT}/lib/intel64 -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_lp64 -lpthread -lm #SIESTA needs an F90 interface to MPI #This will give you SIESTA's own implementation #If your compiler vendor offers an alternative, you may change #to it here. MPI_INTERFACE=libmpi_f90.a MPI_INCLUDE= . #Dependency rules are created by autoconf according to whether #discrete preprocessing is necessary or not. .F.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_fixed_F) $< .F90.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_free_F90) $< .f.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_fixed_f) $< .f90.o: $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_free_f90) $< -- Then I just changed COMP_LIBS=dc_lapack.a to COMP_LIBS += libsiestaLAPACK.a to compile the Siesta trunk 572. It seems it almost finished but an error shows up at the bottom (error message shown below,"gfortran: sockets.o: No such file or directory"). Could you please suggest what happen and how to fix that? Thanks. ... ... mpifc -c -g -O2`FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/m_monitor.F90 mpifc -c -g -O2`FoX/FoX-config --fcflags` /home/tl439/~siesta-maint/siesta/trunk/Src/nag.f mpifc -c -g -O2`FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/pxf.F90 mpifc -c -g -O2`FoX/FoX-config --fcflags` -DFC_HAVE_FLUSH -DFC_HAVE_ABORT -DMPI /home/tl439/~siesta-maint/siesta/trunk/Src/siesta.F mpifc -o siesta \ automatic_cell.o atom_options.o arw.o atomlwf.o bands.o basis_enthalpy.o bessph.o bonds.o born_charge.o cellxc_mod.o cgwf.o chkdim.o chkgmx.o chempot.o coceri.o coxmol.o cross.o compute_norm.o denmat.o denmatlomem.o detover.o dfscf.o m_diagon_opt.o diagon.o digcel.o fft.o dhscf.o constr.o diagk_file.o diagg.o diagk.o diagkp.o diag2g.o diag2k.o diag3g.o diag3k.o diagpol.o diagsprl.o dipole.o dismin.o dnaefs.o doping_uniform.o dot.o m_efield.o egandd.o ener3.o ener3lomem.o errorf.o extrapolon.o m_fixed.o interpolation.o gradient.o gradientlomem.o grdsam.o hsparse.o idiag.o initatom.o inver.o iodm_netcdf.o iodmhs_netcdf.o iogrid_netcdf.o iolwf.o iozm.o ipack.o iopipes.o iosockets.o iowfs_netcdf.o kgrid.o kgridinit.o kinefsm.o ksv.o ksvinit.o linpack.o local_DOS.o madelung.o mesh.o meshphi.o meshdscf.o memory.o meshsubs.o metaforce.o m_supercell.o mulliken.o minvec.o naefs.o m_new_dm.o normalize_dm.o ordern.o outcell.o outcoor.o m_fft_gpfa.o paste.o pdos.o pdosg.o pdosk.o pdoskp.o phirphi.o pixmol.o plcharge.o projected_DOS.o propor.o m_partial_charges.o randomg.o reclat.o redcel.o reinit.o reord.o reordpsi.o rhoofd.o rhoofdsp.o rhooda.o savepsi.o shaper.o timer_tree.o timer.o vmb.o vmat.o vmatsp.o volcel.o cdiag.o rdiag.o cgvc.o cgvc_zmatrix.o m_convergence.o iocg.o ioeig.o iofa.o iokp.o iomd.o kpoint_pdos.o typecell.o ofc.o poison.o readsp.o radfft.o write_md_record.o kpoint_grid.o find_kgrid.o proximity_check.o state_init.o siesta_move.o setup_hamiltonian.o compute_dm.o mixer.o scfconvergence_test.o post_scf_work.o state_analysis.o write_subs.o siesta_init.o struct_init.o siesta_options.o read_options.o siesta_geom.o siesta_analysis.o siesta_end.o siesta_forces.o io.o sparse_matrices.o coor.o atm_transfer.o broadcast_basis.o eggbox.o dsyevds.o zheevds.o optical.o phirphi_opt.o reoptical.o transition_rate.o initparallel.o show_distribution.o setspatial.o setatomnodes.o uncell.o cart2frac.o obc.o precision.o sys.o m_cell.o files.o spatial.o parallel.o parallelsubs.o parsing.o chemical.o atom.o atmparams.o m_mpi_utils.o fdf_extra.o m_iorho.o atmfuncs.o listsc.o memoryinfo.o m_memory.o sorting.o atomlist.o atm_types.o old_atmfuncs.o radial.o alloc.o spher_harm.o periodic_table.o version.o timestamp.o basis_types.o xml.o pseudopotential.o basis_specs.o basis_io.o ldau.o ldau_specs.o onmod.o
Re: [SIESTA-L] Vibra doesn't work for large supercell
Dear All, I think I found what's wrong. the input file for vibra should be the file to build supercell, I used supercell coordinates in the vibra input. After I changed it to unit cell which was used for fcbuild, it works. Sorry about that. Regards, Tao On 2016-07-28 12:44, T. Liu wrote: By looking at the error message ' forrtl: severe (24): end-of-file during read, unit 11, file /home3/e495/e495/tl439/siesta-4.0b-485/Util/Vibra/Src/big/H2O.FC', It seems that there is some problem with .FC file. I think H2O.FC is correct because of right number of row. I should tell that I calculate the .FC in four parts and then 'cat 1.FC 2.FC 3.FC 4.FC > H2O.FC' after deleting the first row of 2.FC 3.FC 4.FC Could anyone suggest what's wrong with this calculation. Thanks. Regards, Tao On 2016-07-28 11:49, T. Liu wrote: Dear All, I am using Vibra to calculate the Phonon frequency for ice (supercell includes 648 atoms). But meet a short error 'Killed' Here is how I did it, Please suggest me what to do. Starting from ice unit cell (24 atoms), use fcbuild build a supercell including 24*3*3*3 = 648 atoms (supercell_1=1 supercell_2=1, supercell_3=1), calculate the Force constants by only displacing the atom in the unit cell, get a .FC file with =648*6*24 +1 = 93313 rows. modify maxa = 648, maxx/y/z=3, recompile vibra, run it and get a very short error message 'killed' if I change maxx/y/z=1, then the following error shows up. recoor: 63713.74701 17.64102 17.82023 1 recoor: 63811.61543 19.85108 17.82135 1 recoor: 63922.20267 16.98088 20.72229 1 recoor: 64014.43389 20.51019 20.71991 1 recoor: 64112.68131 12.97636 13.19783 2 recoor: 64218.31772 18.74695 13.14676 2 recoor: 64318.31898 12.97755 18.91589 2 recoor: 64412.68097 18.74550 18.96677 2 recoor: 64515.49943 15.88767 16.05524 2 recoor: 64621.13456 21.60495 16.05808 2 recoor: 64721.13570 15.83576 21.82704 2 recoor: 64815.49918 21.65692 21.82454 2 lxmax= 0 lymax= 0 lzmax= 0 Number of unit cells in Supercell = 1 Eigenvectors = False Computing Eigenvalues only forrtl: severe (24): end-of-file during read, unit 11, file /home3/e495/e495/tl439/siesta-4.0b-485/Util/Vibra/Src/big/H2O.FC Image PCRoutineLine Source vibra 00436214 Unknown Unknown Unknown vibra 0045BD40 Unknown Unknown Unknown vibra 00404EE7 Unknown Unknown Unknown vibra 00402FDE Unknown Unknown Unknown libc.so.6 2B182C36 Unknown Unknown Unknown vibra 00402EE9 Unknown Unknown Unknown Is there any restrictions on max number of atoms for Vibra? what should I do? I also tested a small cell (unit cell only, 24 atoms) and it works if maxa = 648, maxx/y/z=1. If I just change to maxx/y/z=3, it gives me 'Killed'. If I use maxa=24, and it works for both maxx/y/z=3 and maxx/y/z=1. It seems that there is some restriction on maxasc? Hope I have described enough clear. Regards, Tao
Re: [SIESTA-L] Vibra doesn't work for large supercell
By looking at the error message ' forrtl: severe (24): end-of-file during read, unit 11, file /home3/e495/e495/tl439/siesta-4.0b-485/Util/Vibra/Src/big/H2O.FC', It seems that there is some problem with .FC file. I think H2O.FC is correct because of right number of row. I should tell that I calculate the .FC in four parts and then 'cat 1.FC 2.FC 3.FC 4.FC > H2O.FC' after deleting the first row of 2.FC 3.FC 4.FC Could anyone suggest what's wrong with this calculation. Thanks. Regards, Tao On 2016-07-28 11:49, T. Liu wrote: Dear All, I am using Vibra to calculate the Phonon frequency for ice (supercell includes 648 atoms). But meet a short error 'Killed' Here is how I did it, Please suggest me what to do. Starting from ice unit cell (24 atoms), use fcbuild build a supercell including 24*3*3*3 = 648 atoms (supercell_1=1 supercell_2=1, supercell_3=1), calculate the Force constants by only displacing the atom in the unit cell, get a .FC file with =648*6*24 +1 = 93313 rows. modify maxa = 648, maxx/y/z=3, recompile vibra, run it and get a very short error message 'killed' if I change maxx/y/z=1, then the following error shows up. recoor: 63713.74701 17.64102 17.82023 1 recoor: 63811.61543 19.85108 17.82135 1 recoor: 63922.20267 16.98088 20.72229 1 recoor: 64014.43389 20.51019 20.71991 1 recoor: 64112.68131 12.97636 13.19783 2 recoor: 64218.31772 18.74695 13.14676 2 recoor: 64318.31898 12.97755 18.91589 2 recoor: 64412.68097 18.74550 18.96677 2 recoor: 64515.49943 15.88767 16.05524 2 recoor: 64621.13456 21.60495 16.05808 2 recoor: 64721.13570 15.83576 21.82704 2 recoor: 64815.49918 21.65692 21.82454 2 lxmax= 0 lymax= 0 lzmax= 0 Number of unit cells in Supercell = 1 Eigenvectors = False Computing Eigenvalues only forrtl: severe (24): end-of-file during read, unit 11, file /home3/e495/e495/tl439/siesta-4.0b-485/Util/Vibra/Src/big/H2O.FC Image PCRoutineLine Source vibra 00436214 Unknown Unknown Unknown vibra 0045BD40 Unknown Unknown Unknown vibra 00404EE7 Unknown Unknown Unknown vibra 00402FDE Unknown Unknown Unknown libc.so.6 2B182C36 Unknown Unknown Unknown vibra 00402EE9 Unknown Unknown Unknown Is there any restrictions on max number of atoms for Vibra? what should I do? I also tested a small cell (unit cell only, 24 atoms) and it works if maxa = 648, maxx/y/z=1. If I just change to maxx/y/z=3, it gives me 'Killed'. If I use maxa=24, and it works for both maxx/y/z=3 and maxx/y/z=1. It seems that there is some restriction on maxasc? Hope I have described enough clear. Regards, Tao
[SIESTA-L] Vibra doesn't work for large supercell
Dear All, I am using Vibra to calculate the Phonon frequency for ice (supercell includes 648 atoms). But meet a short error 'Killed' Here is how I did it, Please suggest me what to do. Starting from ice unit cell (24 atoms), use fcbuild build a supercell including 24*3*3*3 = 648 atoms (supercell_1=1 supercell_2=1, supercell_3=1), calculate the Force constants by only displacing the atom in the unit cell, get a .FC file with =648*6*24 +1 = 93313 rows. modify maxa = 648, maxx/y/z=3, recompile vibra, run it and get a very short error message 'killed' if I change maxx/y/z=1, then the following error shows up. recoor: 63713.74701 17.64102 17.82023 1 recoor: 63811.61543 19.85108 17.82135 1 recoor: 63922.20267 16.98088 20.72229 1 recoor: 64014.43389 20.51019 20.71991 1 recoor: 64112.68131 12.97636 13.19783 2 recoor: 64218.31772 18.74695 13.14676 2 recoor: 64318.31898 12.97755 18.91589 2 recoor: 64412.68097 18.74550 18.96677 2 recoor: 64515.49943 15.88767 16.05524 2 recoor: 64621.13456 21.60495 16.05808 2 recoor: 64721.13570 15.83576 21.82704 2 recoor: 64815.49918 21.65692 21.82454 2 lxmax= 0 lymax= 0 lzmax= 0 Number of unit cells in Supercell = 1 Eigenvectors = False Computing Eigenvalues only forrtl: severe (24): end-of-file during read, unit 11, file /home3/e495/e495/tl439/siesta-4.0b-485/Util/Vibra/Src/big/H2O.FC Image PCRoutineLine Source vibra 00436214 Unknown Unknown Unknown vibra 0045BD40 Unknown Unknown Unknown vibra 00404EE7 Unknown Unknown Unknown vibra 00402FDE Unknown Unknown Unknown libc.so.6 2B182C36 Unknown Unknown Unknown vibra 00402EE9 Unknown Unknown Unknown Is there any restrictions on max number of atoms for Vibra? what should I do? I also tested a small cell (unit cell only, 24 atoms) and it works if maxa = 648, maxx/y/z=1. If I just change to maxx/y/z=3, it gives me 'Killed'. If I use maxa=24, and it works for both maxx/y/z=3 and maxx/y/z=1. It seems that there is some restriction on maxasc? Hope I have described enough clear. Regards, Tao
Re: [SIESTA-L] supercell
Thanks, Nick. On 2016-07-27 08:31, Nick Papior wrote: Currently you cannot perform supercell expansions using off-diagonal components. 2016-07-27 9:23 GMT+02:00 T. Liu <tl...@cam.ac.uk>: Dear All, I am trying to use 'supercell' block to make a 2 0 1 0 1 0 0 0 2 supercell, but Siesta stopped and showed that Not safe to use non-diagonal supercells Stopping Program from Node: 0 Program aborted. Backtrace: #0 0x10e968729 #1 0x10e9693a4 #2 0x10ea3e7a8 #3 0x10da4a74a #4 0x10d9160fa #5 0x10d906a39 #6 0x10d904e15 #7 0x10db02bce Abort trap: 6 Could you please tell me if it is possible to make non-diagonal supercell by 'supercell' block? It seems no more detail in the manual. Regards, Tao -- Kind regards Nick
[SIESTA-L] supercell
Dear All, I am trying to use 'supercell' block to make a 2 0 1 0 1 0 0 0 2 supercell, but Siesta stopped and showed that Not safe to use non-diagonal supercells Stopping Program from Node:0 Program aborted. Backtrace: #0 0x10e968729 #1 0x10e9693a4 #2 0x10ea3e7a8 #3 0x10da4a74a #4 0x10d9160fa #5 0x10d906a39 #6 0x10d904e15 #7 0x10db02bce Abort trap: 6 Could you please tell me if it is possible to make non-diagonal supercell by 'supercell' block? It seems no more detail in the manual. Regards, Tao
[SIESTA-L] Inertia momentum of the molecule
Dear All, Is it possible calculate the inertia momentum of the molecule along the unit vectors or x,y,z axis and write the data into output file? Thanks. Regards, Tao