Re: [SIESTA-L] phonon calculations

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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

[T. Liu]

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