0
I Use This!
Moderate Activity

Commits : Listings

Analyzed about 1 hour ago. based on code collected about 1 hour ago.
Dec 18, 2024 — Dec 18, 2025
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
Cleanup NVTX labels More... about 1 year ago
Optimize memory usage in transform_generalized and transform_back_generalized: get rid of tmp/tmpDev array, reuse q/qDev and a/aDev instead More... about 1 year ago
Set default codebranches for generalized eigenproblem: pxgemm for GPU, cannons or hermitian_multiply for CPU as before More... about 1 year ago
Fix allocation/deallocation error More... about 1 year ago
Optimize pxgemm TN/NT: eliminate internal loop for GEMM More... about 1 year ago
Minor bugfix-2: in elpa_pxgemm for no-MPI + GPU More... about 1 year ago
Merge branch 'master_pre_stage' into merge_systems_gpu More... about 1 year ago
Solve for AMD GPUs More... about 1 year ago
Enable only random matrix for multiply tests More... about 1 year ago
Minor bugfix: in elpa_pxgemm for no-MPI + GPU More... about 1 year ago
Minor bugfix: C++ interface for elpa_pxgemm More... about 1 year ago
Add missing elpa_pxgemm_helpers_template.F90 to EXTRA_DIST More... about 1 year ago
Eliminate more unneeded allocating and gpu_host_register'ing for useCCL in cholesky More... about 1 year ago
Eliminate unneeded allocating and gpu_host_register'ing for useCCL case in invert_trm More... about 1 year ago
Merge remote-tracking branch 'origin/peter_nonhermitian_multiply' More... about 1 year ago
Bugfix pxgemm: no-MPI codepath More... about 1 year ago
Bugfix pxgemm: for TN/NT nonsquare-grid GPU codepath More... about 1 year ago
Bugfix pxgemm: for TN/NT nonsquare-grid CPU codepath More... about 1 year ago
Bugfix pxgemm: for TN/NT square-grid CPU codepath More... about 1 year ago
Bugfix: treat correctly TT-case without NCCL in pxgemm_d_ptr More... about 1 year ago
Eliminate unneeded allocating and gpu_host_register'ing of a_tmp for d_ptr case with cu/roc solver in cholesky More... about 1 year ago
Implement a_h_a and d_ptr interfaces for elpa_generalized_eigenvectors and elpa_generalized_eigenvalues More... about 1 year ago
Found and eliminated another sychronization issue. More... about 1 year ago
Merge branch 'peter_fix_elpa1_gpu_codebranches' into 'master_pre_stage' More... about 1 year ago
Cleanup elpa_pxgemm_multiply from USE_CCL_PXGEMM More... about 1 year ago
Found synchro bug at the end of solve_tridi. More... about 1 year ago
Add keyword gpu_pxgemm_multiply More... about 1 year ago
Fix bug for hermitian_multiply device pointer codepath, clean it up More... about 1 year ago
GPU version of solve_tridi_col More... about 1 year ago
Runs with all ELPA components on GPU with MPI Allreduce workaround. Makes memcpy_async sync as a workaround until i find the culprit. More... about 1 year ago