Search found 275 matches

by gmatteo
Wed Mar 16, 2016 12:04 am
Forum: Many-body perturbation theory
Topic: non local part of the pseudo potential ..
Replies: 2
Views: 3522

Re: non local part of the pseudo potential ..

Use inclvkb = 0

The documentation mentions it. Note that inclvkb = 2 is much faster than inclvkb = 1 (the default value in Abint 7.x)
In Abinit8, inclvkb defaults to 2
by gmatteo
Thu Mar 10, 2016 10:37 am
Forum: Many-body perturbation theory
Topic: why no ixc in BSE inputs??
Replies: 1
Views: 2901

Re: why no ixc in BSE inputs??

Because the XC kernel is not needed in the standard BSE formulation. Moreover the value of ixc is automatically detected by the code from the info reported in the header of the pseudopotential files. If you want to perform a BSE calculation with KS wavefunctions and eigenvalues computed with a diffe...
by gmatteo
Thu Mar 10, 2016 10:26 am
Forum: Many-body perturbation theory
Topic: GW nkptgw variable bigger than one possible? [SOLVED]
Replies: 2
Views: 5238

Re: GW nkptgw variable bigger than one possible? [SOLVED]

If you change the value of nkptgw in the input file, you should also modify bdgw.
This variable is an array of shape (2, nkptgw) specifying the range of bands treated with GW
for each k-point in kptgw:

nkptgw 3
bdgw
56 57
56 57
56 57
by gmatteo
Fri Feb 12, 2016 2:20 pm
Forum: Many-body perturbation theory
Topic: BSE tutorial error [SOLVED]
Replies: 3
Views: 5406

Re: BSE tutorial error [SOLVED]

The new version of the GW/BSE code does not use the KSS file anymore. The wavefunctions and the energies are always obtained from the WFK file. This means that you have to create a link to the WKF file produced by tbs_1: So use, ln -s tbs_1o_DS3_WFK tbs_2i_WFK instead of the old command: ln -s tbs_1...
by gmatteo
Tue Feb 02, 2016 6:43 pm
Forum: Ground state
Topic: alloy material
Replies: 2
Views: 2395

Re: alloy material

why occopt 8?
by gmatteo
Mon Jan 18, 2016 11:38 am
Forum: Configuring and compiling ABINIT
Topic: enable-etsf-io [SOLVED]
Replies: 2
Views: 5051

Re: enable-etsf-io [SOLVED]

Dear Natalie, Use --with-trio-flavor="etsf_io+netcdf" to enable ETSF-IO and netcdf support. Please note that there's a problem in the specification of the link flags used for netcdf and etsf-io. One should specify the name of the directory containing the libraries after the -L option (-Ldi...
by gmatteo
Wed Jan 13, 2016 7:21 pm
Forum: Ground state
Topic: Ecut Convergence and Parallelism [SOLVED]
Replies: 2
Views: 4623

Re: Ecut Convergence and Parallelism [SOLVED]

The test on the number of bands was not implemented correctly when paral_kgb == 1.
You can ignore the warning reported in the log file.
by gmatteo
Sat Jan 09, 2016 2:07 pm
Forum: Configuring and compiling ABINIT
Topic: Install Abinit on Redhat HPC [SOLVED]
Replies: 2
Views: 6655

Re: Install Abinit on Redhat HPC [SOLVED]

Your MPI library does not support MPI_INTEGER16.
To make abinit compile:

1) Open the config.h file in your editor and remove the line:
#define FC_INT_QUAD 1

2) make clean
3) make -j4
by gmatteo
Sat Jan 09, 2016 1:56 pm
Forum: Configuring and compiling ABINIT
Topic: Version of PLASMA library supported by Abinit
Replies: 4
Views: 6369

Re: Version of PLASMA library supported by Abinit

Change the interface of abi_xgemm_2d in abi_xgemm.f90 by adding the target attribute to A, B, C === modified file 'src/28_numeric_noabirule/abi_xgemm.f90' --- src/28_numeric_noabirule/abi_xgemm.f90 2015-01-24 06:09:47 +0000 +++ src/28_numeric_noabirule/abi_xgemm.f90 2016-01-09 12:46:39 +0000 @@ -61,...
by gmatteo
Thu May 07, 2015 11:28 am
Forum: Improving ABINIT
Topic: runtests.py failure
Replies: 2
Views: 4051

Re: runtests.py failure

Hi Joe,

have you tried to regenerate the pickle database?
Use

./runtests.py -r -i 'rfddk'
by gmatteo
Tue May 05, 2015 12:38 pm
Forum: Improving ABINIT
Topic: _HIST file
Replies: 6
Views: 6546

Re: _HIST file

didn't look at Matteo's modifs (yes, sometimes, I don't spy!) but I suppose you have stored these time-independent data once for all in the HIST file and not at every time step... Is it the case ? Yes A good starting point could be APPA to list all these needed quantities; they are not numerous... ...
by gmatteo
Tue May 05, 2015 10:49 am
Forum: Improving ABINIT
Topic: _HIST file
Replies: 6
Views: 6546

Re: _HIST file

Hi Jordan, I've already implemented this modification in my 7.11 branch but the code must be merged by Xavier in trunk. Before doing so, however, I have to remove all the calls to the etsf-io library used to write the data in netcdf format since, as we agreed in Liege, Abinit8 won't rely on etsf-io ...
by gmatteo
Wed Jun 25, 2014 2:00 am
Forum: Link to other codes
Topic: question about metaGGA pseudopotential
Replies: 3
Views: 4327

Re: question about metaGGA pseudopotential

Hope the meta-GGA can be performed in the future release of Abinit. Is there any plan about it ?


I'm not aware of any plan about meta-GGA with PAW
by gmatteo
Sun Jun 22, 2014 12:43 am
Forum: Many-body perturbation theory
Topic: SCGW using optdriver8 in abinti-7.6.4 not allowed [SOLVED]
Replies: 2
Views: 5043

Re: SCGW using optdriver8 in abinti-7.6.4 not allowed [SOLVED]

optdriver 8 has been removed in abinit 7.6. The code was difficult to maintain and besides
one can easy implement this kind of algorithm through some kind of driver written
in a high-level language such as python
Sorry for that
by gmatteo
Sun Jun 22, 2014 12:38 am
Forum: Link to other codes
Topic: question about metaGGA pseudopotential
Replies: 3
Views: 4327

Re: question about metaGGA pseudopotential

Is the metagga not implemented in paw in ABINIT? No, meta-GGA with PAW is not available I also trie to generate pseudopotential using APE code. But maybe due to the problem of APE code itself (I found the same error i met in m ... 00018.html), I failed to generate TB09 pseudopotential. Are there ot...
by gmatteo
Tue Jun 17, 2014 10:32 pm
Forum: Configuring and compiling ABINIT
Topic: Unable to compile in parallel using mvapich [SOLVED]
Replies: 6
Views: 6165

Re: Unable to compile in parallel using mvapich [SOLVED]

[quote] So, basicaly i just commented two calls for MPI2 and copy/paste those for MPI1. If it wouldn't corrupt my calculations in future, I say that the problem is SOLVED. Is it? [/quote] Yes, the compilation error due to the missing MPI functions is solved. Note however, that your MPI library is ou...
by gmatteo
Tue Jun 17, 2014 10:28 pm
Forum: Configuring and compiling ABINIT
Topic: error compiling abinit 7.6.2 with external dependencies [SOLVED]
Replies: 6
Views: 7904

Re: error compiling abinit 7.6.2 with netcdf [SOLVED]

"_sfftw_execute_dft_", referenced from: ___m_fftw3_MOD_fftw3_fftpad_spc._omp_fn.15 in lib52_fft_mpi_noabirule.a(m_fftw3.o) ___m_fftw3_MOD_fftw3_fftpad_spc._omp_fn.14 in lib52_fft_mpi_noabirule.a(m_fftw3.o) ___m_fftw3_MOD_fftw3_fftpad_spc._omp_fn.13 in lib52_fft_mpi_noabirule.a(m_fftw3.o) ...
by gmatteo
Tue Jun 17, 2014 1:14 am
Forum: Improving ABINIT
Topic: trying and failing to add a new test [SOLVED]
Replies: 2
Views: 5523

Re: trying and failing to add a new test [SOLVED]

Hi Joe, By default the script runs the entire test suite if not argument is provided. This is handy because otherwise one should specify some kind of option e.g. `runtests.py --all` or an argument e.g. `runtests.py all` to run all the tests. This is indeed a possible solution but on the other hand I...
by gmatteo
Mon Jun 16, 2014 8:08 pm
Forum: Configuring and compiling ABINIT
Topic: Unable to compile in parallel using mvapich [SOLVED]
Replies: 6
Views: 6165

Re: Unable to compile in parallel using mvapich [SOLVED]

Enable the calls to MPI_Errhandler_get and comment the calls to MPI_comm_get_Errhandler in m_xmpi.F90 in subroutine xmpi_comm_set_errhandler === modified file 'src/12_hide_mpi/m_xmpi.F90' --- src/12_hide_mpi/m_xmpi.F90 2014-05-12 09:22:46 +0000 +++ src/12_hide_mpi/m_xmpi.F90 2014-06-16 18:06:01 +000...
by gmatteo
Wed May 07, 2014 1:03 am
Forum: Configuring and compiling ABINIT
Topic: Segmentation fault wannier 90 abinit 7.6.3 [SOLVED]
Replies: 7
Views: 8277

Re: Segmentation fault wannier 90 abinit 7.6.3 [SOLVED]

I have the feeling that the problem is due to VECLIB (Apple's version of BLAS and LAPACK). Could you run wannier.x in the GNU debugger gdb: # To start the debugger: $ gdb path_to_abinit # To start the applications: gdb_prompt> run < testin_wannier90.files ... SEGMENTATION FAULT ... # Use the backtra...
by gmatteo
Wed Apr 30, 2014 2:20 am
Forum: Configuring and compiling ABINIT
Topic: SCF not converged (but really is) parallel BUG [SOLVED]
Replies: 3
Views: 5586

Re: SCF not converged (but really is) parallel BUG [SOLVED]

Changing the -fp-model to strict fixed it.


Where have you changed the compilation option, the MPI library, Abinit or both?
by gmatteo
Wed Apr 23, 2014 1:50 am
Forum: Improving ABINIT
Topic: fldiff
Replies: 5
Views: 11270

Re: fldiff

Hi guys, I already have a preliminary version of fldiff written in python indeed. I'm still working on it but I will commit the python script to my bzr branch asap so that you can look at it and send me comments/suggestions. I wrote the new script with the following goals in mind: ⋅  The c...
by gmatteo
Sun Apr 20, 2014 12:13 am
Forum: Ground state
Topic: mpirun has exited due to process rank 4 abinit-7.6.3
Replies: 7
Views: 6995

Re: mpirun has exited due to process rank 4 abinit-7.6.3

Disable OMP: enable_openmp="no" and recompile the code from scratch: make clean && make -j4 You are linking against the sequential version of MKL thus it does not make sense to enable OpenMP only in abinit. Rerun the calculation and let me know if this solves the problem. I cannot ...
by gmatteo
Sat Apr 19, 2014 11:44 pm
Forum: Ground state
Topic: mpirun has exited due to process rank 4 abinit-7.6.3
Replies: 7
Views: 6995

Re: mpirun has exited due to process rank 4 abinit-7.6.3

I am not using OPENMP Threads;just : Actually you are because you are using enable_openmp="yes" in the configuration file. Set the number of threads to 1 with: $ export OMP_NUM_THREADS=1 Increase the stack size with $ ulimit -s unlimited and try to rerun the calculation with mpirun
by gmatteo
Sat Apr 19, 2014 11:16 pm
Forum: Ground state
Topic: mpirun has exited due to process rank 4 abinit-7.6.3
Replies: 7
Views: 6995

Re: mpirun has exited due to process rank 4 abinit-7.6.3

Could you post the input file and the log file of the run?

How many MPI processes and how many OpenMP threads are you using?
Note that if the env variables OMP_NUM_THREADS is not defined, the OMP runtime library
will use all the CPUS available!