1
0
mirror of https://github.com/paboyle/Grid.git synced 2026-06-21 19:23:17 +01:00

Commit Graph

  • 049cc518f4 Hadrons: introduction message 2 portelli 2018-06-25 19:08:39 +01:00
  • 2e1c66897f Hadrons: introduction message portelli 2018-06-25 19:08:22 +01:00
  • adcef36189 Hadrons: Möbius DWF action portelli 2018-06-25 15:58:35 +01:00
  • 2f121c41c9 Commiting reation of meson field code before a merge with the upstream branch feature/hadrons fionnoh 2018-06-25 12:20:46 +01:00
  • e0ed7e300f Hadrons: spurious Dminus removed portelli 2018-06-22 16:33:43 +02:00
  • 485207901b Merge branch 'develop' into feature/hadrons portelli 2018-06-22 16:15:32 +02:00
  • c760f0a4c3 Hadrons: remove make_5D/4D functions and FreeProp fix portelli 2018-06-22 16:12:46 +02:00
  • c84eeedec3 Hadrons: GaugeProp module for z-Wilson actions portelli 2018-06-22 15:53:22 +02:00
  • 1ac3526f33 Small changes to the A2A header and module fionnoh 2018-06-22 12:29:42 +01:00
  • 0de090ee74 Temporarily added in the contraction code that produced the working 2-pt function. This is commited for reference only and will be removed in the next push. fionnoh 2018-06-22 12:28:41 +01:00
  • 91405de3f7 Hadrons: new solver exposing fermion matrix and generic source/solve import/export portelli 2018-06-22 12:14:37 +02:00
  • 8fccda301a Fixed a bug where the guess was always subtracted after the solve and included appropriate weights for the sources in the one case we're looking at now. More work needs to be done to make the 5d/4d source logic less brittle. fionnoh 2018-06-21 16:36:59 +01:00
  • 7a0abfac89 Restructured the class that computes and returns the A2A vectors. fionnoh 2018-06-21 16:36:06 +01:00
  • ae37fda699 A more elegant way to subtract guesses from solve and a bool check before verifying residual fionnoh 2018-06-20 16:07:40 +01:00
  • b5fc5e2030 All to all module update that hit a promising milestone. Commiting for a reference for future changes. fionnoh 2018-06-20 10:59:07 +01:00
  • cc5d025ea4 WilsonMG: Adapt staggered GMRES/MR tests to "new" constructor Daniel Richtmann 2018-06-18 16:18:56 +02:00
  • 6c97a6a071 Coalescing version of the kernel paboyle 2018-06-13 20:52:29 +01:00
  • 73bb2d5128 Ugly hack to speed up compile on GPU; we don't use the hand kernels on GPU anyway so why compile paboyle 2018-06-13 20:35:28 +01:00
  • b710fec6ea Gpu code first version of specialised kernel paboyle 2018-06-13 20:34:39 +01:00
  • b2a8cd60f5 Doubled gauge field is useful paboyle 2018-06-13 20:27:47 +01:00
  • 867ee364ab Explicit instantiation hooks paboyle 2018-06-13 20:27:12 +01:00
  • 25becc9324 GPU tweaks for benchmarking; really necessary? paboyle 2018-06-13 20:26:07 +01:00
  • 94d1ae4c82 Some prep work for GPU shared memory. Need to be careful, as will try GPU direct RDMA and inter-GPU memory sharing on SUmmit later paboyle 2018-06-13 20:24:06 +01:00
  • 2075b177ef CUDA_ARCH more carefule treatment paboyle 2018-06-13 20:22:34 +01:00
  • 847c761ccc Move sfw IEEE fp16 into central location paboyle 2018-06-13 20:22:01 +01:00
  • 8287ed8383 New GPU vector targets paboyle 2018-06-13 20:21:35 +01:00
  • e6be7416f4 Use managed memory paboyle 2018-06-13 20:14:00 +01:00
  • 26863b6d95 User Managed memory paboyle 2018-06-13 20:13:42 +01:00
  • ebd730bd54 Adding 2D loops paboyle 2018-06-13 20:13:01 +01:00
  • 066be31a3b Optional GPU target SIMD types; work in progress and trying experiments paboyle 2018-06-13 20:07:55 +01:00
  • 7a4c142955 Add GPU specific simd targets paboyle 2018-06-13 19:55:30 +01:00
  • ddcb53bce2 Merge remote-tracking branch 'upstream/develop' into feature/wilsonmg Daniel Richtmann 2018-06-13 09:50:37 +02:00
  • d1c80e1d46 WilsonMG: Correct years in copyright line Daniel Richtmann 2018-06-13 09:42:29 +02:00
  • dc5b909f99 Merge branch 'develop' of https://github.com/paboyle/Grid into feature/Lanczos feature/Lanczos Chulwoo Jung 2018-06-12 19:11:20 -04:00
  • 855b249f57 Merge branch 'develop' of https://github.com/paboyle/Grid into merge2 Chulwoo Jung 2018-06-12 19:08:06 -04:00
  • c73cc7d354 WilsonMG: Add tests with MG preconditioner running single precision, outer solver running in double Daniel Richtmann 2018-06-11 22:05:18 +02:00
  • 49fdc324a0 WilsonMG: Make MG correctness checks abort on failing tests Daniel Richtmann 2018-06-11 16:50:59 +02:00
  • f32714a2d1 WilsonMG: Make running MG correctness checks optional via commandline Daniel Richtmann 2018-06-11 16:27:52 +02:00
  • 73a955be20 WilsonMG: Move tests for Wilson & WilsonClover into separate files Daniel Richtmann 2018-06-09 17:42:19 +02:00
  • 66b7a0f871 WilsonMG: Move multigrid class to separate file Daniel Richtmann 2018-06-12 15:22:02 +02:00
  • 2ab9d4bc56 WilsonMG: Fix random behavior in GMRES Daniel Richtmann 2018-06-11 15:54:32 +02:00
  • 4f41cd114d WilsonMG: Add a mixed precision version of FGMRES Daniel Richtmann 2018-06-09 16:38:36 +02:00
  • 11c4f5e32c WilsonMG: Provide command line switch for reading in input xml + move default params to constructor of MultiGridParams Daniel Richtmann 2018-06-01 10:36:09 +02:00
  • e9b9550298 WilsonMG: Fix incompatibility with single prec MG in construction of simd layout on coarser grids Daniel Richtmann 2018-06-09 16:21:19 +02:00
  • 7564fedf68 WilsonMG: Set subspace to zero to avoid random behavior Daniel Richtmann 2018-05-23 12:24:25 +02:00
  • 8db0ef9736 Merge pull request #168 from jch1g10/feature/qed-fvol portelli 2018-06-08 20:09:06 +02:00
  • 95d4b46446 Merge branch 'develop' of https://github.com/paboyle/Grid into develop Guido Cossu 2018-06-08 11:30:29 +01:00
  • 251b904a28 Merge branch 'release/ISC-freeze-2' ISC-freeze-2 feature/summit-multigrid feature/summit-lanczos paboyle 2018-06-04 21:09:48 +01:00
  • 5dfd216a34 Better thread safety paboyle 2018-06-04 21:08:44 +01:00
  • 5a112feac3 Merge branch 'release/ISC-freeze-1' ISC-freeze-1 paboyle 2018-06-04 18:49:40 +01:00
  • c2e8d0aa88 Solve g++ problem on the lanczos test paboyle 2018-06-04 18:34:15 +01:00
  • 0fe5aeffbb Merge branch 'feature/hadrons' into feature/qed-fvol James Harrison 2018-06-04 16:59:43 +01:00
  • 7fbc469046 Merge branch 'develop' into feature/hadrons James Harrison 2018-06-04 16:58:30 +01:00
  • bf96a4bdbf Merge branch 'master' into develop paboyle 2018-06-04 14:03:11 +01:00
  • 84685c9bc3 Overflow fix paboyle 2018-06-04 13:42:07 +01:00
  • a8d4156997 Added a Hadrons module that computes the all-to-all v and w vectors fionnoh 2018-05-31 17:18:58 +01:00
  • c18074869b Changes to Hadrons SchurRB solver to allow for a subtract_guess boolean to be passed fionnoh 2018-05-31 17:17:16 +01:00
  • f4c6d39238 CHanges made to SchurRB solvers to allow for the subtraction of a guess after solve fionnoh 2018-05-31 17:16:20 +01:00
  • a5645a7efe for dwf_precondition feature/dwf-preconditioning Azusa Yamaguchi 2018-05-30 09:48:52 +01:00
  • 8b3e738269 for dwf_precondition Azusa Yamaguchi 2018-05-30 09:47:38 +01:00
  • 118746b1e9 Adding Mobius BlockCG test Chulwoo Jung 2018-05-28 18:29:50 +00:00
  • 200d35b38a Merge branch 'develop' into feature/hadrons portelli 2018-05-28 11:52:47 +02:00
  • eb52e84d09 Merge branch 'feature/hadrons' of github.com:paboyle/Grid into feature/hadrons portelli 2018-05-28 11:50:27 +02:00
  • 72abc34764 Merge pull request #166 from guelpers/feature/hadrons portelli 2018-05-28 11:49:46 +02:00
  • e3164d4c7b Hadrons: env function to get volume in double portelli 2018-05-28 11:39:17 +02:00
  • 8f6039646b Added Hermition check in BlockCG Chulwoo Jung 2018-05-24 02:56:53 -04:00
  • 95e9fd1889 More diag output Chulwoo Jung 2018-05-23 04:08:21 -04:00
  • f5db386c55 Change MODULE_REGISTER_NS -> MODULE_REGISTER in UnitEM, ScalarVP and VPCounterTerms James Harrison 2018-05-22 16:16:21 +01:00
  • 66da4a38f9 Added Lattice I/O Chulwoo Jung 2018-05-22 00:21:25 -04:00
  • 236868d2e9 Checking in vectorized Blocked CG Chulwoo Jung 2018-05-21 18:51:59 -04:00
  • 294ee70a7a Merge branch 'feature/hadrons' into feature/qed-fvol James Harrison 2018-05-21 18:02:41 +01:00
  • 013ea4e8d1 Merge branch 'feature/staggered-comms-compute' into develop Azusa Yamaguchi 2018-05-21 13:11:56 +01:00
  • 7fbbb31a50 Merge branch 'develop' into feature/staggered-comms-compute feature/staggered-comms-compute Azusa Yamaguchi 2018-05-21 13:07:29 +01:00
  • 0e127b1fc7 New file single prec test Azusa Yamaguchi 2018-05-21 12:57:13 +01:00
  • 68c028b0a6 Comment Azusa Yamaguchi 2018-05-21 12:54:25 +01:00
  • 255d4992e1 Hadrons: stochastic scalar SU(N) free field fix portelli 2018-05-18 20:49:55 +01:00
  • a0d399e5ce Hadrons: yet other attempts at EMT NPR portelli 2018-05-18 20:49:26 +01:00
  • fd3b2e945a Hadrons: don't right result with empty stem portelli 2018-05-18 20:48:24 +01:00
  • 6c27c72585 WilsonMG: Provide more sensible default values for MG parameters Daniel Richtmann 2018-05-16 17:08:05 +02:00
  • 9c003d2d72 WilsonMG: Base wilson mg preconditioner entirely on existing infrastructure Daniel Richtmann 2018-05-16 16:46:25 +02:00
  • 4b8710970c WilsonMG: Switch to Galerkin coarsening in CoarsenedMatrix Daniel Richtmann 2018-05-16 17:23:12 +02:00
  • 68d686ec38 WilsonMG: Add functionality for applying G5 on coarse grids Daniel Richtmann 2018-05-16 16:17:14 +02:00
  • c48b69ca81 WilsonMG: Implement Mdir & Mdiag in CoarsenedMatrix Daniel Richtmann 2018-05-16 16:07:58 +02:00
  • df8c208f5c WilsonMG: Revert CoarsenedMatrix.h and Lattice_transfer.h back to state of develop branch Daniel Richtmann 2018-05-16 16:02:17 +02:00
  • 61812ab7f1 Merge remote-tracking branch 'upstream/develop' into feature/wilsonmg Daniel Richtmann 2018-05-15 14:57:18 +02:00
  • b999984501 Merge branch 'develop' into feature/hadrons portelli 2018-05-15 13:53:57 +01:00
  • 7836cc2d74 No checksum output on log for scidac Guido Cossu 2018-05-15 10:10:08 +01:00
  • eb7d34a4cc GPU version Peter Boyle 2018-05-14 19:41:47 -04:00
  • aab27a655a Start of GPU kernels Peter Boyle 2018-05-14 19:41:17 -04:00
  • 93280bae85 Gpu option Peter Boyle 2018-05-14 19:40:58 -04:00
  • c5f93abcd7 GPU clean up Peter Boyle 2018-05-14 19:40:33 -04:00
  • d5deef782d Useful debug comments Peter Boyle 2018-05-14 19:39:52 -04:00
  • 5f50473c0d Clean up Peter Boyle 2018-05-14 19:39:11 -04:00
  • a61e0df54b Travis fix for Lime portelli 2018-05-14 19:56:12 +01:00
  • 9d835afa35 Attempt at solving the FP exception in the QED code portelli 2018-05-14 19:05:54 +01:00
  • 5e3be47117 Hadrons: scalar SU(N) various fixes portelli 2018-05-14 18:58:39 +01:00
  • 13f50406e3 Suppress print statement Peter Boyle 2018-05-12 18:00:00 -04:00
  • 09cd46d337 Lane by Lane operation Peter Boyle 2018-05-12 17:59:35 -04:00
  • d3f51065c2 Give command line control of blocks/threads split Peter Boyle 2018-05-12 17:58:56 -04:00
  • 925ac4173d Thread count control for warp scheduler thingy doodaa thing Peter Boyle 2018-05-12 17:58:22 -04:00