- 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 9 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
-
Pavel Kus authored
matrices of dimension of the form 2^n * 3^m * 5^m are now allowed. For other matrix sizes, the test is terminated. Matrix size is not modified, as it has been the case before
-
Andreas Marek authored
-
- 28 Jul, 2017 2 commits
-
-
Lorenz Huedepohl authored
There was again a case where stack-variables were loaded with instructions that needed properly aligned memory. This only surfaced with the Intel C compiler, where the stack layout evidently was sufficiently different to trigger this. This was also the case for SSE kernels
-
Andreas Marek authored
- remove runtime by not executing unecessary runs - make sure that on gp02 impi/2017 is NOT use - adjust reserved time on knl systems
-
- 27 Jul, 2017 2 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-
- 25 Jul, 2017 8 commits
-
-
Lorenz Huedepohl authored
-
Lorenz Huedepohl authored
The new Makefile rule just dumped more and more lines into the test_x.sh script, as the ">>" operator was used always.
-
Lorenz Huedepohl authored
-
-
-
-
and moved the implementation to test_util.F90
-
Lorenz Huedepohl authored
Due to the extremely slow compilation on KNL this error now surfaced, as libelpatest.la was not ready in time, there :)
-
- 20 Jul, 2017 3 commits
-
-
Andreas Marek authored
-
Pavel Kus authored
Previously we checked ortonormality of eigenvectors by comparing the matrix S^T * S to identity matrix. The new feature is also checking just the diagonal of the matrix S^T * S. By that we get the information, how far are the eigenvectors from having length 1. If it turns out to be important, we could try to normalize them at the end of elpa, which is simple (in contrast with enforcing better orthogonality).
-
Andreas Marek authored
This stupid bug was there since commit ae45bbb3
-
- 19 Jul, 2017 9 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-
Andreas Marek authored
From now on call make check with TASKS=x, e.g. make check TASKS=2
-
Pavel Kus authored
-
Pavel Kus authored
-
Pavel Kus authored
-
Pavel Kus authored
-
Pavel Kus authored
-
Andreas Marek authored
-
- 18 Jul, 2017 2 commits
-
-
Andreas Marek authored
-
Andreas Marek authored
-