1. 01 Aug, 2018 3 commits
  2. 05 Jun, 2018 1 commit
  3. 29 May, 2018 1 commit
  4. 25 May, 2018 1 commit
  5. 24 May, 2018 1 commit
  6. 22 May, 2018 1 commit
  7. 11 May, 2018 1 commit
  8. 30 Apr, 2018 1 commit
  9. 17 Apr, 2018 1 commit
  10. 14 Mar, 2018 1 commit
  11. 05 Mar, 2018 1 commit
  12. 06 Feb, 2018 1 commit
  13. 05 Feb, 2018 4 commits
  14. 08 Jan, 2018 1 commit
  15. 14 Dec, 2017 2 commits
  16. 01 Dec, 2017 2 commits
  17. 30 Nov, 2017 1 commit
  18. 27 Nov, 2017 1 commit
  19. 25 Nov, 2017 1 commit
  20. 24 Nov, 2017 2 commits
  21. 07 Nov, 2017 1 commit
  22. 30 Oct, 2017 1 commit
  23. 26 Oct, 2017 2 commits
    • Lorenz Huedepohl's avatar
      First implementation of an autotuning procedure · 0ebe4802
      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)
      0ebe4802
    • Lorenz Huedepohl's avatar
      67c4a5dd
  24. 21 Aug, 2017 1 commit
  25. 08 Aug, 2017 1 commit
  26. 03 Aug, 2017 1 commit
    • Lorenz Huedepohl's avatar
      Fix memory leak · 72700e47
      Lorenz Huedepohl authored
      The timer object was not properly removed when elpa_destroy() was called
      on an ELPA instance, leading to a memory leak
      72700e47
  27. 30 Jul, 2017 1 commit
    • Lorenz Huedepohl's avatar
      Loop over all possible domain decompositions · fabb1c42
      Lorenz Huedepohl authored
      We got reports from a user that there were troubles with certain domain
      decompositions. So far the tests only looked at (approximately) square
      decompositions in column-major process order.
      
      Now, a new class of tests loops over all possible decompositions
      (row * col) for a given number of total tasks.
      
      So far, we can not confirm that there are any problems, all
      possibilities work as expected.
      fabb1c42
  28. 15 Jul, 2017 1 commit
  29. 07 Jul, 2017 2 commits
  30. 09 Jun, 2017 1 commit
    • Lorenz Huedepohl's avatar
      Prevent some more warnings about uninitialized uses · 6688e40c
      Lorenz Huedepohl authored
      It is not quite clear to me why GCC complains here when
      WANT_SINGLE_PRECISION_REAL is not defined, as the variables are then not
      used _at all_.
      
      Nonetheless, this again prevents build on the SuSE OBS, so fix it to
      make the compiler happy
      6688e40c