elpa merge requestshttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests2021-12-08T12:16:57Zhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/88A100 kernel2021-12-08T12:16:57ZAndreas MarekA100 kernelAndreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/115Add configure flag to enable hipcub2022-10-28T14:19:50ZAndreas MarekAdd configure flag to enable hipcubFor nblk > 64 the HIPCUB implementation gives
wrong results!For nblk > 64 the HIPCUB implementation gives
wrong results!Andreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/110Add configure flag to enable hipcub2022-08-16T06:46:39ZAndreas MarekAdd configure flag to enable hipcubFor nblk > 64 the HIPCUB implementation gives
wrong results!For nblk > 64 the HIPCUB implementation gives
wrong results!Andreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/133Add distcheck-gpu CI test and fix missing GPU source file in Makefile2023-01-30T13:40:15ZPetr KarpovAdd distcheck-gpu CI test and fix missing GPU source file in MakefileAdd missing file test_gpu_vendor_agnostic_layerVariables.h in Makefile and add CI-test "distcheck-gpu"Add missing file test_gpu_vendor_agnostic_layerVariables.h in Makefile and add CI-test "distcheck-gpu"Andreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/121Added C++ interface2022-11-14T12:48:19ZPetr KarpovAdded C++ interfaceAndreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/164Add GPU device information to gpu_object2024-02-23T07:11:26ZAndreas MarekAdd GPU device information to gpu_object- At start up some GPU devices parameters are queried and stored
- For example the count of SM processors is passed to (some) kernels- At start up some GPU devices parameters are queried and stored
- For example the count of SM processors is passed to (some) kernelshttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/15adding test which splits communicator at the beginning2019-05-08T12:40:39ZPavel Kusadding test which splits communicator at the beginningand runs several elpas (still mpi-based) in parallel. Each elpa
"instance" then splits its communicator to row and column communicator,
as it is always done. This is actually often done in FHI-aims and was
previously leading to issues wh...and runs several elpas (still mpi-based) in parallel. Each elpa
"instance" then splits its communicator to row and column communicator,
as it is always done. This is actually often done in FHI-aims and was
previously leading to issues when using OmniPath. Hopefully was solved
by Intel MPI 2019.3. Keeping the test for future reference.https://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/171Add man page for elpa_setup_gpu2024-03-14T08:42:40ZPetr KarpovAdd man page for elpa_setup_gpuAndreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/57Add missing variable to openmp clause shared2021-02-24T12:09:15ZAndreas MarekAdd missing variable to openmp clause sharedAndreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/8Add python interface for ELPA2018-11-14T20:42:10ZSebastian OhlmannAdd python interface for ELPAThe ELPA C API is wrapped by cython code that can be called from python. Python classes for convenience access are provided. Tests are included and activated in CI.
The interface is built when --enable-python is given to configure. The ...The ELPA C API is wrapped by cython code that can be called from python. Python classes for convenience access are provided. Tests are included and activated in CI.
The interface is built when --enable-python is given to configure. The requirements then are then python, numpy, mpi4py, cython. The tests can be enabled with --enable-python-tests; then, an additional requirement is pytest.
Currently, the interface is only tested with the MPI version of ELPA.https://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/22Add support for NVTX profiling2019-10-23T10:03:12ZSebastian OhlmannAdd support for NVTX profilingWhen profiling the GPU version, NVTX can be used to highlight the corresponding regions of the code in the timeline of the profiling tool (nvvp or nsight systems). This is very useful to correlate what happens on the GPU with what part o...When profiling the GPU version, NVTX can be used to highlight the corresponding regions of the code in the timeline of the profiling tool (nvvp or nsight systems). This is very useful to correlate what happens on the GPU with what part of the code we are in.
Currently, the regions are only defined in the elpa1 solver.https://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/132Allow 1 MPI rank for ELPA22023-01-25T10:32:09ZPetr KarpovAllow 1 MPI rank for ELPA2Andreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/108Allow to use NVIDIA cub in real GPU kernel (might give ~10% speedup)2022-08-06T07:03:58ZAndreas MarekAllow to use NVIDIA cub in real GPU kernel (might give ~10% speedup)Andreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/38Allow user code to set threshold for check_pd2020-07-25T07:55:19ZWenzhe YuAllow user code to set threshold for check_pdThis feature is used in FHI-aims and other quantum chemistry and electronic structure codes as a standard approach to handle ill-conditioning. For example:
* [The Journal of Chemical Physics 152, 194103 (2020)](https://doi.org/10.1063/5....This feature is used in FHI-aims and other quantum chemistry and electronic structure codes as a standard approach to handle ill-conditioning. For example:
* [The Journal of Chemical Physics 152, 194103 (2020)](https://doi.org/10.1063/5.0007045)
* [The Journal of Chemical Physics 151, 241102 (2019)](https://doi.org/10.1063/1.5139948)https://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/114Amd mi2502022-10-20T09:28:20ZAndreas MarekAmd mi250Andreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/51Arm NEON2020-12-10T11:36:54ZAndreas MarekArm NEONAndreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/49Arm sve2020-11-30T19:28:30ZAndreas MarekArm sveAndreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/55Assign tasks to gpu2021-02-15T12:49:46ZAndreas MarekAssign tasks to gpuAndreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/159Async in trans ev2023-12-21T08:15:53ZAndreas MarekAsync in trans evAndreas MarekAndreas Marekhttps://gitlab.mpcdf.mpg.de/elpa/elpa/-/merge_requests/20Auto detect2019-10-11T05:47:35ZAndreas MarekAuto detect