dbcsr

dbcsr copied to clipboard

Include leak and thread sanitize checkers

1

Include in GNU tests: - [x] `-fsanitize=leak` - [ ] `-fsanitize=thread`, check https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55561 and CP2K toolchain - [ ] Others?

alazzaro

tests: make libsmm_acc tests run with MPI

1

currently the tests output unconditionally and are not being run through the MPI wrappers (`mpirun` or `srun`) in `tests/CMakeLists.txt`

dev-zero

Daint-CI log not accessible

10

See also [here](https://github.com/cp2k/cp2k/pull/1751#issuecomment-975523824).

hfp

Add CP2K support for A100

1

I am trying to build CP2K for our NVIDIA A100 system using spack, I reported an issue and according to @dev-zero in https://github.com/spack/spack/issues/28554#issuecomment-1033840686 he suggested we post an issue here...

shahzebsiddiqui

In the File `dbcsr_api_c.F` in the subroutine `c_free_string` the intrinsic subroutine `c_f_pointer` is called: `CALL c_f_pointer(c_string, string)`. The second argument is declared as `CHARACTER(:, kind=c_char), POINTER :: string`. The Fortran...

SpinTensor

See https://github.com/cp2k/cp2k/issues/1036

alazzaro

Currently, thread_dist is part the dbcsr_distribution. However, thread_dist is based on row_blk_sizes, which are part of dbcsr_type. So we can have a mix where a matrix has a thread distribution...

alazzaro

See https://github.com/cp2k/cp2k/issues/1037

alazzaro

Inconsistent INTENT for nullified pointer DATA

1

In most test cases at dbcsr_array_types.F line 93 and 155 the pointer DATA is nullified. However in calling procedures the corresponding pointer is an INTENT(IN) formal argument. It is called...

vzecca

Number of threads has changed! when running with cp2k.sdbg and more than 64 OMP threads

5

**Describe the bug** we get the following error message when running certain tests with a fresh `cp2k.sdbg`: ``` ******************************************************************************* * ___ * * / \ * * [ABORT] * *...

dev-zero