sivkov

In my fork/branch I fixed it. You can test https://github.com/falltergeist/falltergeist/commit/d34abb18fbb9d19b5f377225cd4454b5588acd50

not only generation. Also multiplication, I use eigensolver of lapack to test Arnoldi in LA branch

which functions should be tested?

I tried to move creation of the grid inside the creation of mp_env, but tests are crashing with MPI ABORT. Currently I reverted these changes and added manually creation and...

this is my branch an commit where I refactor grid and init_lib https://github.com/dithillobothrium/dbcsr-1/commit/06d8d51da95d507f46bba81df40f0d4489f7b287

it is postponed because of the https://github.com/cp2k/dbcsr/blob/22e274011d9b31bbfd851d6eee71a9d102ce0f98/src/ops/dbcsr_test_methods.F#L401

Sorry, I need to assign someone here @sanshar @gkc1000 @hczhai

> Since you do not have `nonspinadapted` in the input file, this is SU(2) spin-adapted calculation (alpha = beta). > In the input file `spin 1` means total spin `2S...

You mean the final rdm is rdm taken from a mixed state? In this case rdm becomes not very useful, since it is averaged over ensemble. But I don't understand...

Hi, I compiled with boost 1.74 and OMPI 4.0 and it works. But I needed to change some MPI calls in the code.