E

ELPA

Settings | Report Duplicate

0

I Use This!

Moderate Activity

Commits : Listings

Analyzed about 5 hours ago. based on code collected about 5 hours ago.

Commit Message	Contributor	Files Modified	Lines Added	Lines Removed	Code Location	Date
Jun 25, 2024 — Jun 25, 2025 Showing page 5 of 143 Search / Filter on:
Rename AMD sources: .cpp -> .hip, add preprocessing for .hip sources	Petr Karpov	More...				8 months ago
Add preprocessing for .cu sources	Petr Karpov	More...				8 months ago
Minor renaming: useGPU -> do_useGPU in elpa1_template.F90	Petr Karpov	More...				8 months ago
Save-1. Hipifly elpa_pxgemm_multiply kernels -- create a universal cuda/hip interface	Petr Karpov	More...				8 months ago
Split elpa_multiply_a_b_cuda.cu to elpa_hermitian_multiply_cuda.cu and elpa_pxgemm_multiply_cuda.cu (and same for hip)	Petr Karpov	More...				8 months ago
Enable switching on/off GPU branches for individual steps in ELPA1 (tridiag, solve_tridi, trans_ev)	Petr Karpov	More...				8 months ago
Add new key-value pair (runtime option) pxgemm_for_generalized, with default value 1	Petr Karpov	More...				8 months ago
Bugfix for elpa_hermitian_multiply, CPU-codepath	Petr Karpov	More...				8 months ago
Bugfixes for elpa_pxgemm_multiply complex datatype	Petr Karpov	More...				8 months ago
Add explicit interfaces and delete unused parameters uplo_a, uplo_c for elpa_pxgemm_multiply	Petr Karpov	More...				8 months ago
Add tests for non-hermitian (non-symmetric) matrices for elpa_pxgemm	Petr Karpov	More...				8 months ago
Bugfix: pxgemm, square-grid GPU codepath	Petr Karpov	More...				8 months ago
Add pxgemm HIP kernels	Petr Karpov	More...				8 months ago
Add HIP, SYCL, OpenMP offload fortran interfaces for pxgemm	Peter Karpov	More...				8 months ago
Switch on square-grid codepaths in pxgemm	Peter Karpov	More...				8 months ago
Switch on elpa_pxgemm_multiply instead of elpa_hermitian_multiply for generalized eigenproblem	Petr Karpov	More...				8 months ago
Create d_ptr codepath for pxgemm-NN,TT	Petr Karpov	More...				8 months ago
Cleanup NVTX labels in multiply_a_b/elpa_pxgemm_a_b_template.F90	Petr Karpov	More...				8 months ago
pxgemm-TN,NT: cleanup and optimize d_ptr codepath	Petr Karpov	More...				8 months ago
Bugfix pxgemm: introduce CPU buffer for no-send op in elpa_transpose_row_or_col, non-NCCL	Petr Karpov	More...				8 months ago
Bugfix pxgemm: reduce max number of threads to 512 in cuda_update_c_tn_nt_kernel	Petr Karpov	More...				8 months ago
Port pxgemm-TN,NT nonsquare-grid codepath to NCCL	Petr Karpov	More...				8 months ago
Port update_c to GPU in pxgemm	Petr Karpov	More...				8 months ago
Merge branch 'origin/master_pre_stage' into peter_nonhermitian_multiply	Petr Karpov	More...				9 months ago
Initial port of pxgemm-TN,NT nonsquare-grid coepath to GPU	Petr Karpov	More...				9 months ago
Merge branch 'peter_ubuntu_cholesky_fix' into 'master_pre_stage'	Andreas Marek	More...				9 months ago
Fix single-precision case	Andreas Marek	More...				9 months ago
Increase the error check limit for single precision Cholesky from 0.09 to 0.3 to silence the problem for builds with Reference LAPACK on Ubuntu	Peter Karpov	More...				9 months ago
Implement pxgemm-NT nonsquare-grid CPU codepath	Petr Karpov	More...				9 months ago
Fix error in non-stream case	Andreas Marek	More...				9 months ago

←
1
2
3
4
5
6
7
8
9
…
142
143
→