Commits · master · elpa / elpa

08 Aug, 2022 1 commit
- Merge branch 'master_pre_stage' into 'master' · be044ee3
  Andreas Marek authored Aug 08, 2022
```
If user does not set "omp_threads", then...

See merge request !109
```
  be044ee3
04 Aug, 2022 1 commit
- Merge branch 'AMD-Mi250' into 'master_pre_stage' · f5aed623
  Andreas Marek authored Aug 04, 2022
```
Improvements for AMD Mi250

See merge request !107
```
  f5aed623
03 Aug, 2022 2 commits
- new cholesky kernels needs debuging · ffeb4c33
  Andreas Marek authored Aug 03, 2022
  
  ffeb4c33
- Fix error in check scripts for srun wrapper · 11d009ad
  Andreas Marek authored Aug 03, 2022
  
  11d009ad
02 Aug, 2022 8 commits
- Allow to specify a MPI launcher · f9405ec9
  Andreas Marek authored Aug 02, 2022
  
  f9405ec9
- Reactivate all CI tests · df9fd868
  Andreas Marek authored Aug 02, 2022
  
  df9fd868
- Revert HIP cholesky to previous version which works · 3dff3024
  Andreas Marek authored Aug 02, 2022
  
  3dff3024
- Prefer srun over mpiexec as MPI launcher · f570c9f8
  Andreas Marek authored Aug 02, 2022
  
  f570c9f8
- remove debug print in elpa_invert_trm_hip · 635f4a08
  Andreas Marek authored Aug 02, 2022
  
  635f4a08
- Also tests need to work with HIP on Nvidia · 4a01303f
  Andreas Marek authored Aug 02, 2022
  
  4a01303f
- Configure flags for HIP on NVIDIA · f8bd0238
  Andreas Marek authored Aug 02, 2022
  
  f8bd0238
- CI test for HIP on NVIDIA · 79f5a92b
  Andreas Marek authored Aug 02, 2022
  
  79f5a92b
01 Aug, 2022 1 commit
- Make AMD Code port with HIP on Nvidia GPUs · ea366d25
  Andreas Marek authored Aug 01, 2022
  
  ea366d25
28 Jul, 2022 1 commit
- Fix error in using 'int' in HostRegister · 308d6cc4
  Andreas Marek authored Jul 28, 2022
  
  308d6cc4
25 Jul, 2022 1 commit
- Fix typos in AMD stack · 2b0c0300
  Andreas Marek authored Jul 25, 2022
  
  2b0c0300
21 Jul, 2022 3 commits
- If user does not set "omp_threads", then... · 60217f90
  Andreas Marek authored Jul 20, 2022
```
use the value specified by omp_get_max_threads()
```
  60217f90
- some comments · d188ad0a
  Andreas Marek authored Jul 21, 2022
  
  d188ad0a
- Fix error in checking of stream_synchronize · 0b0b59c9
  Andreas Marek authored Jul 21, 2022
  
  0b0b59c9
15 Jul, 2022 1 commit
- Merge branch 'florian_bndred_blocking2' into AMD-Mi250 · bed76e42
  Andreas Marek authored Jul 15, 2022
  
  bed76e42
13 Jul, 2022 1 commit
- AMD GPU improvements provided by AMD · accadb2f
  Andreas Marek authored Jul 13, 2022
  
  accadb2f
01 Jul, 2022 1 commit
- Trigger all CI tests · b040342c
  Andreas Marek authored Jul 01, 2022
  
  b040342c
30 Jun, 2022 2 commits
- Use MPI_STATUSES_IGNORE in mpi_waitall · 63f89d5b
  Andreas Marek authored Jun 30, 2022
  
  63f89d5b
- Update the CI compilers · bdd3184c
  Andreas Marek authored Jun 30, 2022
  
  bdd3184c
29 Jun, 2022 17 commits
- another GPU fix · 432a6136
  Florian Merz authored Jun 29, 2022
  
  432a6136
- fixes for GPU · f5850510
  Florian Merz authored Jun 29, 2022
  
  f5850510
- some cleanup · c51ff99a
  Florian Merz authored Jun 28, 2022
  
  c51ff99a
- fixed openmp · 46242cf4
  Florian Merz authored Jun 28, 2022
  
  46242cf4
- do bcasts in single go · cd8b4816
  Florian Merz authored Jun 27, 2022
  
  cd8b4816
- block structure in subroutine · 7d280156
  Florian Merz authored Jun 23, 2022
  
  7d280156
- introduced blocked loop · 80b231e1
  Florian Merz authored Jun 23, 2022
  
  80b231e1
- small fix for long int · d331fb60
  Florian Merz authored Jun 03, 2022
  
  d331fb60
- use unblocked version for openmp and gpus for now · 4fb30d1e
  Florian Merz authored Jun 03, 2022
  
  4fb30d1e
- moved storing of Householder vectors out of first loop · 5bf97762
  Florian Merz authored May 30, 2022
  
  5bf97762
- removed tmat dependency in first loop · 52e8d134
  Florian Merz authored May 30, 2022
  
  52e8d134
- fixed sign error from previous commit · fc2b29dc
  Florian Merz authored May 29, 2022
  
  fc2b29dc
- Make no-mpi compile; still errors at runtime · ac9a7789
  Andreas Marek authored May 26, 2022 and Florian Merz committed Jun 29, 2022
  
  ac9a7789
- Make ilp64 compile; still errors at runtime · 835a3045
  Andreas Marek authored May 26, 2022 and Florian Merz committed Jun 29, 2022
  
  835a3045
- fixes for complex/single precision · 031ecaa5
  Florian Merz authored May 17, 2022
  
  031ecaa5
- fixes after rebase · 6e215722
  Florian Merz authored May 16, 2022
  
  6e215722
- more blas in hh_block · 9f2c9cc9
  Florian Merz authored May 10, 2022
  
  9f2c9cc9