- 02 Mar, 2020 1 commit
-
-
Andreas Marek authored
-
- 24 Oct, 2019 1 commit
-
-
Andreas Marek authored
-
- 14 Oct, 2019 1 commit
-
-
Andreas Marek authored
ELPA can now link agains a 64bit integer verion of BLAS/LAPACK. Currently this only works if ELPA is compiled with MPI=OFF! The 64bit support is not available in the legacy interface
-
- 11 Oct, 2019 2 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-
- 23 Sep, 2019 1 commit
-
-
Andreas Marek authored
-
- 03 Jul, 2019 1 commit
-
-
Andreas Marek authored
-
- 24 Jun, 2019 1 commit
-
-
Andreas Marek authored
-
- 18 Apr, 2019 1 commit
-
-
Andreas Marek authored
This solves issue #66
-
- 23 Nov, 2018 1 commit
-
-
Andreas Marek authored
-
- 21 Nov, 2018 1 commit
-
-
Pavel Kus authored
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)
-
- 15 Nov, 2018 1 commit
-
- 13 Nov, 2018 1 commit
-
-
Andreas Marek authored
-
- 10 Aug, 2018 1 commit
-
-
Pavel Kus authored
-
- 08 Aug, 2018 1 commit
-
-
Pavel Kus authored
-
- 12 Jun, 2018 1 commit
-
-
Andreas Marek authored
-
- 22 May, 2018 1 commit
-
-
Pavel Kus authored
on selected mpi ranks using suppress_warnings option
-
- 18 Apr, 2018 3 commits
- 05 Feb, 2018 5 commits
-
-
Pavel Kus authored
instead, blacs context has to be set beforehand
-
Pavel Kus authored
added possibility to re-use the already decomposed matrix B in the generalized problem Added respective test
-
Pavel Kus authored
* tested with elpa1 only (should be sufficient) * not using Hermitian multiply at the moment * requires scalapack
-
Pavel Kus authored
-
Pavel Kus authored
and some modifications in test.f90
-
- 27 Nov, 2017 1 commit
-
-
Andreas Marek authored
-
- 07 Nov, 2017 3 commits
-
-
Pavel Kus authored
-
Pavel Kus authored
-
Pavel Kus authored
This test was wrong, it was computing A * A^T instead of A^T * A. The latter is correct since our implementation of Cholesky decomposition stores the triangular matrix in the upper triangle The test was passing only because the Cholesky decomposition was tested with diagonal matrix only, than this does not matter.
-
- 30 Oct, 2017 1 commit
-
-
Pavel Kus authored
-
- 26 Oct, 2017 1 commit
-
-
Lorenz Huedepohl authored
To be used like this class(elpa_t), pointer :: e class(elpa_autotune_t), pointer :: tune_state e => elpa_allocate() call e%set(...) [...] assert_elpa_ok(e%setup()) tune_state => e%autotune_setup(ELPA_AUTOTUNE_FAST, ELPA_AUTOTUNE_DOMAIN_REAL) ! Autotuning loop, continues until all combinations have been tried do while (e%autotune_step(tune_state)) ! Do the steps that are representative of your calculation call e%eigenvectors(a, ev, z, error) end do ! Fix best parameters, and de-allocate the autotune object call e%autotune_set_best(tune_state) call elpa_autotune_deallocate(tune_state)
-
- 25 Oct, 2017 3 commits
- 11 Sep, 2017 2 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
But allow this in cholesky and hermitian multiply test
-
- 10 Sep, 2017 1 commit
-
-
Andreas Marek authored
-
- 09 Sep, 2017 2 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
It is planned to add another matrix type for the tests. The names of the prepare routines have become a bit inconsistent and confusing. Thus the rename
-
- 03 Sep, 2017 1 commit
-
-
Pavel Kus authored
unlike elpa 2stage, scalapack computes only part of eigenvalues if only part of eigenvectors are required. Check only those.
-