Commits · 760a1b4876ba549788f8209ecf7479b98e12edcd · Giacomo Mulas / NP_TMcode · GitLab

Sep 23, 2025
- Use omp parallel for loops for embrrassingly parallel sections of LUCIN · 760a1b48
  Giovanni La Mura authored Sep 23, 2025
  
  760a1b48
- Use vectorized version of AM in LUCIN · 77e57978
  Giovanni La Mura authored Sep 23, 2025
  
  77e57978
Sep 22, 2025
- Have cdtp() working on the vectorized version of AM · f0d4acb7
  Giovanni La Mura authored Sep 22, 2025
  
  f0d4acb7
Sep 19, 2025
- Use 0-indexed vectorized AM in LUCIN · 6cea13bc
  Giovanni La Mura authored Sep 19, 2025
  
  6cea13bc
- Make sure that the logger is destroyed independently of whether code is under MPI or not · ab8f3c2d
  Giovanni La Mura authored Sep 19, 2025
  
  ab8f3c2d
Sep 18, 2025
- Merge branch 'script_devel' into 'master' · e39cb1e4
  Giovanni La Mura authored Sep 18, 2025
  e39cb1e4
- Trapping grid scale · 98851a54
  Giovanni La Mura authored Sep 18, 2025
  
  98851a54
Sep 16, 2025
- Merge branch 'optimize_trapping' into 'master' · 07859038
  Giovanni La Mura authored Sep 16, 2025
  07859038
- Improve parsing scripts · e939fed7
  Giovanni La Mura authored Sep 16, 2025
  
  e939fed7
Sep 12, 2025
- Merge branch 'script_devel' into 'master' · 59f29e3d
  Giovanni La Mura authored Sep 12, 2025
  59f29e3d
- Update commit logging · 8de79ba5
  Giovanni La Mura authored Sep 12, 2025
  
  8de79ba5
Sep 05, 2025
- Merge branch 'optimize_trapping' into 'master' · c1fe34a6
  Giacomo Mulas authored Sep 05, 2025
  c1fe34a6
- Enable XI scale interpreter for np_cluster · 25d51758
  Giovanni La Mura authored Sep 05, 2025
  
  25d51758
- Implement np_sphere output parser · b7431d86
  Giovanni La Mura authored Sep 05, 2025
  
  b7431d86
Sep 04, 2025
- Implement np_cluster legacy result parser · 5e3b1843
  Giovanni La Mura authored Sep 04, 2025
  
  5e3b1843
- Create output parsing script · cde330cd
  Giovanni La Mura authored Sep 04, 2025
  
  cde330cd
Aug 29, 2025
- Use long long indices also in CPU parallel loops · 7289e202
  Giovanni La Mura authored Aug 29, 2025
  
  7289e202
Aug 12, 2025
- Use lon long indices on large loops and vector access · a3f8befc
  Giovanni La Mura authored Aug 12, 2025
  
  a3f8befc
Aug 08, 2025
- Revise FRFME offloaded loop implementation · 3e0025ff
  Giovanni La Mura authored Aug 08, 2025
  
  3e0025ff
- Use aligned vector allocations when offloading to GPU · 4dfc33af
  Giovanni La Mura authored Aug 08, 2025
  
  4dfc33af
- Make TFRFME::vec_wsum directly accessible · 57ac3b1a
  Giovanni La Mura authored Aug 08, 2025
  
  57ac3b1a
- Make Swap2::vec_vkzm directly accessible · 0becb9e3
  Giovanni La Mura authored Aug 08, 2025
  
  0becb9e3
- Use complex exponential function in CPU exclusive code · 0f705d1e
  Giovanni La Mura authored Aug 08, 2025
  
  0f705d1e
- Use CPU to initialize global work space and result vector, then run a single GPU kernel · d081ed92
  Giovanni La Mura authored Aug 08, 2025
  
  d081ed92
Aug 07, 2025
- Flatten j80 loop and add logging on number of J iterations · a4af2237
  Giovanni La Mura authored Aug 07, 2025
  
  a4af2237
Aug 06, 2025
- Use a target data region with multiple kernels instead of separate functions with a target context · 43d18f76
  Giovanni La Mura authored Aug 06, 2025
  
  43d18f76
- Use sincos() instead of separate calls to sin() and cos() · e940395a
  Giovanni La Mura authored Aug 06, 2025
  
  e940395a
- Merge branch 'offload_trapping' into 'master' · 120a5b3a
  Giovanni La Mura authored Aug 06, 2025
  120a5b3a
- Bracket nvtxRange calls into compiler flags · 3af28bf7
  Giovanni La Mura authored Aug 06, 2025
  
  3af28bf7
- Remove intervening variable declarations in collapsed loops (unsupported by old compilers) · f2018157
  Giovanni La Mura authored Aug 06, 2025
  
  f2018157
- Comment out offload dependent code in clu_subs · 44000bfd
  Giovanni La Mura authored Aug 06, 2025
  
  44000bfd
- Use pre-computed indices in loop and fix timers · 71c48282
  Giovanni La Mura authored Aug 06, 2025
  
  71c48282
Aug 05, 2025
- Flatten offloaded loop and add stage-dependent benchmark timers · 5795b9a1
  Giovanni La Mura authored Aug 05, 2025
  
  5795b9a1
- Use a runtime configuration option to skip FRFME if already computed · 16e0d698
  Giovanni La Mura authored Aug 05, 2025
  
  16e0d698
- Prevent parsing error if runtime options are too short · f17ebcb1
  Giovanni La Mura authored Aug 05, 2025
  
  f17ebcb1
Aug 04, 2025
- Refactor the offloaded nested loop structure to take advantage from teams · 730486ef
  Giovanni La Mura authored Aug 04, 2025
  
  730486ef
- Fix nested loop linearization for GPU and CPU codes · a0a3f870
  Giovanni La Mura authored Aug 04, 2025
  
  a0a3f870
Jul 31, 2025
- Fix offload mapping of cfrfme loop · 17d8bd16
  Giovanni La Mura authored Jul 31, 2025
  
  17d8bd16
Jul 22, 2025
- Add vkv and vec_vkzm vectors to mapping · 5d3eb583
  Giovanni La Mura authored Jul 22, 2025
  
  5d3eb583
Jul 21, 2025
- Added doxygen inline documentation expansion · cb308b85
  Giovanni La Mura authored Jul 21, 2025
  
  cb308b85