- 10 Dec, 2018 1 commit
-
-
Pavel Kus authored
by allreducing the measured time and using average over ranks insted
-
- 04 Dec, 2018 1 commit
-
-
Pavel Kus authored
-
- 28 Nov, 2018 2 commits
- 23 Nov, 2018 1 commit
-
-
Andreas Marek authored
-
- 21 Nov, 2018 3 commits
-
-
Pavel Kus authored
-
Pavel Kus authored
the error handling has to be thought of properly in some of the uses of check_elpa_get/set (e.g. those outside the elpa_setup function)
-
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)
-
- 20 Nov, 2018 3 commits
-
-
Andreas Marek authored
-
-
-
- 15 Nov, 2018 1 commit
-
-
Pavel Kus authored
-
- 14 Nov, 2018 3 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
C interfaces added for: - elpa_print_settings - elpa_store_settings - elpa_load_settings - elpa_autotune_print_state Tests are done for: - elpa_autotune_print_best - elpa_autotune_print_state - elpa_print_settings There are _NO_ tests at the moment for the load/store settings C routine Also the C interfaces for load/save autotuning are still missing
-
Andreas Marek authored
-
- 13 Nov, 2018 4 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
- 07 Nov, 2018 1 commit
-
-
Pavel Kus authored
-
- 27 Aug, 2018 3 commits
- 14 Aug, 2018 3 commits
- 13 Aug, 2018 2 commits
- 10 Aug, 2018 2 commits
- 08 Aug, 2018 1 commit
-
-
Pavel Kus authored
-
- 01 Aug, 2018 3 commits
- 05 Jun, 2018 1 commit
-
-
Andreas Marek authored
-
- 29 May, 2018 1 commit
-
-
Andreas Marek authored
-
- 25 May, 2018 1 commit
-
-
Lorenz Huedepohl authored
Lead to the first configuration be tested twice
-
- 24 May, 2018 1 commit
-
-
Lead to the first configuration be tested twice
-
- 22 May, 2018 1 commit
-
-
Pavel Kus authored
time_spent is not initialized. Should be dealt with in a better way
-
- 11 May, 2018 1 commit
-
-
Andreas Marek authored
-