GitLab

This solves issue #66

to avoid stalling between storing and loading parameters

checking that the loaded values equal to the stored values

mpi_comm_parent is allways requred (it was not required before, but
actually the code internals expected it to be supplied, at least for
ELPA 2 calculation OR whenever GPU was used)

by putting #ifdef WIHT_MPI where it should have been

Pavel Kus authored Nov 09, 2018 and Pavel Kus committed Nov 15, 2018

and supplied to the elpa object by set methods

this commit does not work, the generalized EVP problems fail from 2
reasons:
- Cannons algorithm does not work for the processor grid shape we use at
  the moment in tests (It only works if num. of process columns is a
  multiple of num. of process rows, which is not true in ELPA for 2 mpi
  tasks)
- Cannons algorithm does not work as integrated to ELPA for larger
  matrices (not clear why at the moment)

There are 2 new tests, which should work
- test_cannon.c: it tests the new algorithm without going through
  Fortran, as it has been delivered
- test_c_bindings: it tests transfering a 2D fortran matrix to C and
  back, as it is done with the cannons algorithm in the normal ELPA
  tests

on selected mpi ranks using suppress_warnings option

avoiding calling both check eigenvalues AND check eigenvalues and
eigenvectors for analytic eigenvector test

if provided a a fourth parameter, all correctness checks will be
skipped. Can be usefull for repeated performance tests.

instead, blacs context has to be set beforehand

added possibility to re-use the already decomposed matrix B in the
generalized problem
Added respective test

* tested with elpa1 only (should be sufficient)
* not using Hermitian multiply at the moment
* requires scalapack

and some modifications in test.f90