ABINIT Discussion Forums

Dear Davide,

Thank you for reaching out and for the patch. From what I see, this is already fixed in the development branch of ABINIT.

FYI, the --enable-netcdf-default option is only used for testing purposes and will be removed from the upcoming ABINIT 9.0.

As parf of the Minisymposium `Bringing Scientific Applications Written in Fortran to the Exascale Era: How Software Engineering Can Help to Fill the Gap` at the next PASC19, we prepared a survey on Fortran: m Would it be possible for you to share within your collaborators (or could you suggest some...

Please note that GSL support is about to be removed because it is highly marginal and unmaintained. The only call to it occurs in src/32_util/m_special_funcs.F90.

You can safely leave it disabled.

In order to know which options the configure script of ABINIT accepts, you just have to run:

./configure --help

from the top ABINIT source directory.

I'll fix this ASAP. This will be released in the next version of ABINIT. In the meantime, if you have the Autotools installed on your system, you can edit the following files: * config/hints/cc_intel_default-hints.conf * config/hints/fc_intel_default-hints.conf and run config/scripts/makemake before...

FYI: The bugfix for this problem is currently in the pipes and will be made available with ABINIT 8.8.3. Thank you for your patience.

A temporary measure could be to rebuild ABINIT with an older version of GCC.

You should use the option:

If you have the Autotools (autoconf, automake, libtool) installed on your computer, you can edit config/m4/conn-dft.m4 and change line 216 from: ]])], [abi_dft_libxc_version="yes"], [abi_dft_libxc_version="no"] to: ]])], [abi_dft_libxc_version="yes"], [abi_dft_libxc_ver...

In progress. You can follow the MR from the Abinit Gitlab: https://gitlab.abinit.org/trunk/abinit/ ... quests/174

I've reported this issue in the bug tracker of the Abinit Gitlab. You can follow its evolution at: https://gitlab.abinit.org/trunk/abinit/issues/19

We are seeking to fill a postdoctoral position opened at Centro de Física de Materiales (m) at the group of Prof. Daniel Sánchez-Portal (m). The work will focus on the optimization and development of the Ehrenfest dynamics (real-time TDDFT) options on SIESTA (in collaboration with the group of Prof....

Abinit bindings to V_Sim are currently broken and unmaintained. They might be replaced by something else in the near future. There will be developers meeting this year where we will decide what to do with this part of the code. We will inform you when a decision has been taken.

If you have access to a graphical interface based on Unity or Gnome, go to System settings > Software & Updates. You will see a "Download from" field. Choose an appropriate server there.

You are using a prehistoric 4.1 version of GCC. Please upgrade your compilers to at least version 4.8.

Mac OS is a bit picky about library order. Just reverse "-lxc -lxcf90" to "-lxcf90 -lxc" and it should work.

Abinit 8.0 is compatible with LibXC 2.2. Unless you need a very recently added functional, you should use this version.

If you do need LibXC 3.0, please wait for the upcoming Abinit 8.2.

Just install NetCDF on your computer and make sure you have the Fortran module. It should be in /usr/include after installation.

This happens because you are using an old version of GCC which is not supported anymore by the build system. Please upgrade your GCC to version 4.8 or above. We currently recommend GCC 5.4 and are working on fully supporting GCC 6. In the meantime, you can specify optimization flags when running con...

There has obviously been a mistake when generating the source tarball, with the use of an outdated version of the Autotools. We will take care of generating the future releases of all Abinit components with an up-to-date development environment. In the meantime, since you are not using any optional ...

This is a problem with your MPI installation, not with Abinit. Please consult their documentation / forums / mailing lists and/or re-install MPI.

The patched version of Wannier90 is hosted there:

https://launchpad.net/wannier90

We have identified a few bugs in the detection of linear algebra, that will be fixed in the upcoming 8.0 and 8.2 versions of Abinit. In the meantime, you may try the following: ../configure [your options] \ CPPFLAGS_EXTRA="-DHAVE_LINALG_ELPA -DHAVE_LINALG_ELPA_2015" \ FPPFLAGS_EXTRA="...

This problem is likely not related to Abinit. Looks like LibXC doesn't like the space character in your install path. Please rename "Abinit instalation" to "Abinit_instalation" and retry.

If it works, please report the issue to the developers of LibXC.

I wouldn't worry too much about GPU issues, because they will bring an interesting speed-up only in specific cases. If your hardware is a bit outdated, MPI and OpenMP will do a pretty good job already.

Achieving the fastest speed will depend a lot on your architecture and on the kind of calculations you perform. There is no generic recipe that would guarantee better performance. And in some situtations, better performance means also less stability. You have to try different combinations. On an Int...