1
0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-07 01:47:07 +01:00

Commit Graph

  • fcf5023845 Running on Frontier Peter Boyle 2023-10-05 16:45:27 -04:00
  • c8adad6d8b First runs on Summit. PopulateAdag needs work Peter Boyle 2023-10-05 16:43:37 -04:00
  • 737d3ffb98 ADEF1 and 1 hop projection Peter Boyle 2023-10-03 14:22:18 -04:00
  • 6d0c2de399 Deprecate teh PVC directory and make a PVC-OEM generic PVC target with no queueing system dependency -- just interactive scripts Peter Boyle 2023-10-03 17:04:20 +00:00
  • 7786ea9921 Bug fix in script Peter Boyle 2023-10-03 09:58:44 -07:00
  • d93eac7b1c Performance regressed and is OK in icpx 2023.2 Peter Boyle 2023-10-03 15:53:14 +00:00
  • b01e67bab1 coalescedReadGeneralPermute now working Peter Boyle 2023-10-02 17:46:57 -04:00
  • 8a70314f54 Merge branch 'develop' into feature/scidac-wp1 Peter Boyle 2023-10-02 17:24:55 -04:00
  • afc316f501 Rename headers Peter Boyle 2023-10-02 16:25:11 -04:00
  • f14bfd5c1b Relocate sub includes Peter Boyle 2023-10-02 16:23:38 -04:00
  • c5f1420dea Merge remote-tracking branch 'LupoA/develop' into LupoA-develop Peter Boyle 2023-10-02 16:22:35 -04:00
  • 018e6da872
    Merge pull request #440 from giltirn/feature/paddedcellgauge Peter Boyle 2023-10-02 10:00:42 -04:00
  • b77bccfac2
    Merge pull request #444 from mmphys/feature/docX Peter Boyle 2023-10-02 09:57:11 -04:00
  • 36ae6e5aba Fastest GPU version. Need to work on the PaddedCell now to make much faster Peter Boyle 2023-09-29 18:26:51 -04:00
  • 9db585cfeb Temporary commit while optimisation is carried out Peter Boyle 2023-09-29 17:11:35 -04:00
  • c564611ba7 Annoying hack that is useful to preserve for profiling Peter Boyle 2023-09-29 17:11:12 -04:00
  • e187bcb85c Updating Peter Boyle 2023-09-29 17:10:17 -04:00
  • be18ffe3b4 Further tuning and lanczos Peter Boyle 2023-09-27 16:21:58 -04:00
  • 0d63dce4e2 Timing info Peter Boyle 2023-09-27 16:21:14 -04:00
  • 26b30e1551 Flop count and projection to nearest neighbour (keeps redundant flops) Peter Boyle 2023-09-27 16:20:11 -04:00
  • 7fc58ac293 Verbose subspace init Peter Boyle 2023-09-27 16:19:45 -04:00
  • 3a86cce8c1 Compile Peter Boyle 2023-09-27 16:19:18 -04:00
  • 80359e0d49 Bland SYCL compile Peter Boyle 2023-09-26 13:20:27 -07:00
  • 3d437c5cc4 Making SYCL happy Peter Boyle 2023-09-26 13:19:42 -07:00
  • 37884d369f Coarse space is expensive, but gives a speed up in fine matrix multiplies now. Down to optimisation Peter Boyle 2023-09-25 17:24:19 -04:00
  • 9246e653cd Basic non-local coarsening of operator test Peter Boyle 2023-09-25 17:20:58 -04:00
  • 64283c8673 Normal equations becomes linear function for easy base class pass aroudn Peter Boyle 2023-09-25 17:19:39 -04:00
  • 755002da9c Comparison convenience Peter Boyle 2023-09-25 17:16:33 -04:00
  • 31b8e8b437 Better messaging Peter Boyle 2023-09-25 17:16:14 -04:00
  • 0ec0de97e6 Adef2 implemented and working in an HDCG like context Peter Boyle 2023-09-25 17:15:03 -04:00
  • 6c3ade5d89 Improved the coarsening Peter Boyle 2023-09-25 17:14:40 -04:00
  • 980c5f9a34 Update chebyshev setup Peter Boyle 2023-09-25 17:12:22 -04:00
  • 63d9b8e8a3 Merge remote-tracking branch 'origin/develop' into hisq_fat_links david clarke 2023-09-16 23:20:31 -06:00
  • d247031c98 try 7-link david clarke 2023-09-16 23:18:16 -06:00
  • e29b97b3ea Qslash term added Peter Boyle 2023-09-14 16:14:03 -04:00
  • ad2b699d2b Better macos Peter Boyle 2023-09-14 16:12:21 -04:00
  • 471ca5f281 Power method more iterations Peter Boyle 2023-09-07 10:55:05 -04:00
  • e82ddcff5d Working getting closer to HDCG but some low level engineering work still needed + MUCH work on optimisation Peter Boyle 2023-09-07 10:53:51 -04:00
  • b9dcad89e8 Test cases for coarsening with non-local stencil Peter Boyle 2023-09-07 10:53:22 -04:00
  • 993f43ef4a Even odd use case Peter Boyle 2023-09-07 10:53:06 -04:00
  • 2b43308208 First cut non-local coarsening Peter Boyle 2023-08-25 17:38:07 -04:00
  • 04a1ac3a76 First cut for non-local coarsening Peter Boyle 2023-08-25 17:37:38 -04:00
  • 990b8798bd Merge remote-tracking branch 'refs/remotes/origin/develop' into develop Peter Boyle 2023-08-25 17:36:45 -04:00
  • b334a73a44 Stencil improvement Peter Boyle 2023-08-25 17:35:10 -04:00
  • 5d113d1c70 Odd address sanitizer complain Peter Boyle 2023-08-25 17:34:18 -04:00
  • c14977aeab Random vector option for test purposes Peter Boyle 2023-08-25 17:33:31 -04:00
  • 3e94838204 Spread out improvement Peter Boyle 2023-08-25 17:31:28 -04:00
  • c0a0b8ca62 NEON and address sanitiser Peter Boyle 2023-08-25 17:30:30 -04:00
  • b8a7004365 Partial fraction test Peter Boyle 2023-08-14 15:17:03 -04:00
  • affff3865f Merge branch 'develop' into hisq_fat_links david clarke 2023-08-11 23:08:04 -06:00
  • 9c22655b5a Merge remote-tracking branch 'origin/develop' into develop david clarke 2023-08-11 23:06:42 -06:00
  • 99d879ea7f 5-link first attempt david clarke 2023-08-11 22:56:30 -06:00
  • bd56c95a6f Update documentation with complete list of Macports needed to build Grid on a fresh Mac Michael Marshall 2023-07-14 13:50:06 +01:00
  • 994512048e
    Merge pull request #439 from felixerben/bugfix/IRL_convergence Peter Boyle 2023-07-12 16:32:26 -04:00
  • dbd8bb49dc
    Merge pull request #32 from LupoA/sp2n/develop chillenzer 2023-07-04 15:23:43 +00:00
  • 3a29af0ce4 Fixed linker error Julian Lenz 2023-07-04 16:08:44 +01:00
  • f7b79cdd45 Added test for ProjectSpn Julian Lenz 2023-07-03 18:00:32 +01:00
  • 075b9d22d0 adjoint rep implemented as 2indx symmetric Alessandro Lupo 2023-07-02 13:58:31 +01:00
  • b92428f05f better test Alessandro Lupo 2023-07-02 13:34:03 +01:00
  • 34b11864b6 prettiest tests Alessandro Lupo 2023-07-02 13:25:57 +01:00
  • 1dfaa08afb The stencils for the staple and rect-staple padded cell implementations are now created and stored by workspace classes that allow for reuse providing the grids remain consistent The workspaces are now used by the plaq+rectangle gauge action resulting in a further 2x performance improvement as measured on a 16^4 local volume for 2 nodes (16 ranks) of Crusher Christopher Kelly 2023-06-28 15:11:24 -04:00
  • 9d263d9a7d fix bug in HISQSmearing; move benchmark b/c i don't understand how makefiles work david clarke 2023-06-28 10:05:34 -06:00
  • 9015c229dc add benchmark to see whether matrix multiplication is slower than read from object david clarke 2023-06-27 21:28:26 -06:00
  • f44dce390f Implemented acclerator-optimized versions of localCopyRegion and insertSliceLocal to speed up padding Fixed const correctness on PaddedCell methods Fixed compile issues on Crusher Added timing breakdowns for PaddedCell::Expand and the padded implementations of the staples, visible under --log Performance Optimized kernel for StaplePadded Test_iwasaki_action_newstaple now repeats the calculation 10 times and reports average timings Christopher Kelly 2023-06-27 14:58:10 -04:00
  • bb71e9a96a Added PaddedCell and GeneralisedLocalStencil header includes to standard base headers Moved versions of the padded-cell implementations of staple and rect-staple from test code to WilsonLoops header Added StapleAndRectStapleAll which is now called by the plaq+rectangle action class. Under the hood it uses the padded cell implementations with maximal reuse of the padded gauge links Christopher Kelly 2023-06-27 11:23:30 -04:00
  • 78bae9417c returning Nstop vectors even if not all meet true convergence criterion Felix Erben 2023-06-27 14:38:19 +01:00
  • dd170ead01 whitespace Felix Erben 2023-06-27 11:37:01 +01:00
  • 014704856f do one more iteration if not all vectors converged Felix Erben 2023-06-27 11:33:30 +01:00
  • a7eabaad56 rudimentary appendShift convenience method, which allows the user to append an arbitrary shift in one line david clarke 2023-06-26 23:59:28 -06:00
  • eeb4703b84 develop wrappers to make the stencils easier to construct david clarke 2023-06-26 17:45:35 -06:00
  • a07421b3d3 Merge branch 'develop' into hisq_fat_links david clarke 2023-06-26 13:51:32 -06:00
  • cda53b4068 Merge remote-tracking branch 'origin/develop' into develop david clarke 2023-06-26 13:51:06 -06:00
  • 6f6844ccf1 Added new StapleAll and RectStapleAll functions that return the staples for all mu as an array Modified plaq+rectangle gauge actions to use the above Added a test code to confirm the above changes Christopher Kelly 2023-06-26 15:48:47 -04:00
  • 4c6613d72c Modified RectStapleDouble and RectStapleOptimised to use Gauge-BC respecting CshiftLink Added test code tests/debug/Test_optimized_staple_gaugebc demonstrating equivalence of above to RectStapleUnoptimised for cconj gauge BCs Removed optimized staple only being used for periodic gauge BCs; it is now always used Christopher Kelly 2023-06-26 10:20:23 -04:00
  • ee92e08edb
    Merge pull request #435 from fjosw/fix/warnings_in_WilsonKernelsImplementation Peter Boyle 2023-06-23 11:47:19 -04:00
  • c1dcee9328
    Merge pull request #437 from fjosw/fix/stencil_debug Peter Boyle 2023-06-23 11:47:00 -04:00
  • 559257bbe9 better documentation and filelist names Alessandro Lupo 2023-06-23 16:16:48 +01:00
  • 6b150961fe Better script Peter Boyle 2023-06-23 18:09:25 +03:00
  • cff1f8d3b8 rm unused variables and formatting Alessandro Lupo 2023-06-23 16:04:18 +01:00
  • f27d2083cd adjustments in SUn and Sp2n impl Alessandro Lupo 2023-06-23 15:34:08 +01:00
  • 36cc9c524f Threaded the constructor of GeneralLocalStencil Christopher Kelly 2023-06-23 09:57:38 -04:00
  • 2822487450 rm unncessary line Alessandro Lupo 2023-06-23 14:55:23 +01:00
  • e07fafe46a minor adjustments to twoindex Alessandro Lupo 2023-06-23 12:18:04 +01:00
  • 063d290bd8 missing function Alessandro Lupo 2023-06-23 11:11:20 +01:00
  • 4e6194d92a Avoid code duplication in ProjectSUn Alessandro Lupo 2023-06-23 11:03:50 +01:00
  • de30c4e22a minor improvements Alessandro Lupo 2023-06-23 10:49:41 +01:00
  • df99f227c1 include missing staple orientations; invert path direction, which was backwards david clarke 2023-06-22 14:57:10 -06:00
  • 5bafcaedfa Merge branch 'develop' of https://github.com/paboyle/Grid into develop Peter Boyle 2023-06-22 19:59:45 +03:00
  • bfeceae708 FTHMC Peter Boyle 2023-06-22 12:58:18 -04:00
  • eacb66591f Config command Peter Boyle 2023-06-22 19:56:40 +03:00
  • fadaa85626 Update Peter Boyle 2023-06-22 19:56:27 +03:00
  • 02a5b0d786 Updating run during testing Peter Boyle 2023-06-22 19:52:46 +03:00
  • 0e2141442a Dennis says broken Peter Boyle 2023-06-22 19:19:51 +03:00
  • 769eb0eecb Precision coverage Peter Boyle 2023-06-22 19:19:20 +03:00
  • 4241c7d4a3 Imported coalescedReadGeneralPermute GPU implementation from Christoph Fixed bug in padded staple code where extract was being called on the result before the GPU view was closed Fixed compile issue with pointer cast in padded staple code Added timing summaries of padded staple code and timing breakdown of staple implementation to Test_padded_cell_staple Christopher Kelly 2023-06-21 16:01:01 -04:00
  • d536c67b9d add HISQSmearing to Smearing.h david clarke 2023-06-20 16:04:48 -06:00
  • f44f005dad rename _lvl1 --> _linkTreatment david clarke 2023-06-20 15:48:27 -06:00
  • 26b2caf570 add template parameter to Smear_HISQ_fat for MILC interfacing david clarke 2023-06-20 15:37:54 -06:00
  • 7b11075102 The user can now specify the implementation of Cshift used by the PaddedCell class through a virtual base class API. Implementations for default (regular Cshift) and for gauge links (which respects the gauge BCs) Fixed const-correctness for PaddedCell and ConjugateGimpl::setDirections Modified test code for padded-cell implementation of staple, rect-staple to use cconj BCs Christopher Kelly 2023-06-20 17:09:56 -04:00
  • abc658dca5 Added coalescedReadGeneralPermute CPU implementation based on Christoph's GPT code In a test code, implemented a padded-cell version of the staple and rectangular-staple calculation Christopher Kelly 2023-06-20 16:14:25 -04:00