- 10 Aug, 2017 3 commits
-
-
Pavel Kus authored
yet another attempt to improve single/double and real/complex unification more readable
-
Lorenz Huedepohl authored
with obvious meaning
-
Pavel Kus authored
for easier comparisons of elpa and mkl, a test case using scalapack function pdsyevd has been added
-
- 08 Aug, 2017 4 commits
-
-
Andreas Marek authored
-
Lorenz Huedepohl authored
The MPI communicators were never freed, this created a memory leak for every elpa_allocate()/elpa_deallocate() cycle.
-
Andreas Marek authored
-
Andreas Marek authored
-
- 05 Aug, 2017 5 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
- 04 Aug, 2017 2 commits
-
-
Andreas Marek authored
-
Lorenz Huedepohl authored
The were a couple of places where a load of a stack variable was too wide. Detected with the -fsanitize=address flag of GCC, that we should probably incorporate into our CI tests.
-
- 03 Aug, 2017 12 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Lorenz Huedepohl authored
Anything if it makes Andreas happy :)
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Lorenz Huedepohl authored
The timer object was not properly removed when elpa_destroy() was called on an ELPA instance, leading to a memory leak
-
Lorenz Huedepohl authored
Previously you could set up ELPA with nev > na and force a crash.
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
- 02 Aug, 2017 4 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
- 01 Aug, 2017 4 commits
-
-
Pavel Kus authored
even if only part of eigenvectors are calculated, all eigenvalues should be checked
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
- 31 Jul, 2017 2 commits
-
-
Andreas Marek authored
-
Lorenz Huedepohl authored
-
- 30 Jul, 2017 3 commits
-
-
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.
-
Andreas Marek authored
-
Andreas Marek authored
-
- 29 Jul, 2017 1 commit
-
-
Andreas Marek authored
-