| 54e9436 | paboyle | 24 June 2017, 22:10:24 UTC | Experimental: Multiple communicators to see if we can avoid thread locks in --enable-comms=mpit | 24 June 2017, 22:10:24 UTC |
| 869b99e | paboyle | 24 June 2017, 09:55:54 UTC | Threaded calls to multiple communicators | 24 June 2017, 09:55:54 UTC |
| 4372d04 | Antonin Portelli | 23 June 2017, 15:59:27 UTC | Merge pull request #118 from Lanny91/hotfix/bgq Hotfix/bgq | 23 June 2017, 15:59:27 UTC |
| 56abbdf | Lanny91 | 23 June 2017, 09:09:14 UTC | AVX512 integer reduce fix (for non-intel compiler) | 23 June 2017, 09:09:14 UTC |
| af71c63 | Lanny91 | 23 June 2017, 09:03:12 UTC | AVX2 fix | 23 June 2017, 09:03:12 UTC |
| 0440d4c | Lanny91 | 22 June 2017, 15:09:42 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into hotfix/bgq | 22 June 2017, 15:09:42 UTC |
| b22eab8 | Antonin Portelli | 21 June 2017, 17:32:48 UTC | Merge commit 'a7d56523abee6c9030fdd9303c79954897b1086f' into feature/hadrons | 21 June 2017, 17:32:48 UTC |
| a7d5652 | paboyle | 21 June 2017, 13:03:20 UTC | Merge branch 'feature/lanczos-simplify' into develop | 21 June 2017, 13:03:20 UTC |
| 9e56c65 | paboyle | 21 June 2017, 13:02:58 UTC | Updated TODO list | 21 June 2017, 13:02:58 UTC |
| ef4f2b8 | paboyle | 21 June 2017, 08:22:20 UTC | todo update | 21 June 2017, 08:22:20 UTC |
| e8b95bd | paboyle | 21 June 2017, 01:50:09 UTC | Clean up finished. Could shrink Lanczos to around 400 lines at a push | 21 June 2017, 01:50:09 UTC |
| 7e35286 | paboyle | 21 June 2017, 01:26:03 UTC | Simplified lanczos, added Eigen diagonalisation. Curious if we can deprecate dependencly on BLAS. Will see when we get 48^3 running on our BG/Q port | 21 June 2017, 01:26:03 UTC |
| 0486ff8 | paboyle | 20 June 2017, 17:46:01 UTC | Improved the lancos | 20 June 2017, 17:46:01 UTC |
| 1e8a2e1 | Antonin Portelli | 20 June 2017, 16:24:55 UTC | various compatibility fixes after merge | 20 June 2017, 16:24:55 UTC |
| 7587df8 | Antonin Portelli | 20 June 2017, 14:50:39 UTC | Merge branch 'develop' into feature/hadrons # Conflicts: # lib/qcd/action/scalar/ScalarImpl.h | 20 June 2017, 14:50:39 UTC |
| e9cc219 | Azusa Yamaguchi | 20 June 2017, 11:37:41 UTC | Block solver complete for staggered. Now stable on mass 0.003 and gives 8x (!) speed up on Haswell laptop vs. standard CG for 8 RHS solves. 166 iterations vs. 537 iterations so algorithmic gain + 2x in flop rate gain. Better than a slap in the face with a wet kipper. | 20 June 2017, 11:37:41 UTC |
| 0a8faac | Azusa Yamaguchi | 19 June 2017, 21:54:18 UTC | Fix make tests compile | 19 June 2017, 21:54:18 UTC |
| abc4de0 | Azusa Yamaguchi | 19 June 2017, 21:03:03 UTC | No compile make tests fix | 19 June 2017, 21:03:03 UTC |
| cfe3cd7 | Azusa Yamaguchi | 19 June 2017, 13:04:21 UTC | Block solver improvements | 19 June 2017, 13:04:21 UTC |
| 3fa5e31 | Azusa Yamaguchi | 19 June 2017, 13:01:44 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 19 June 2017, 13:01:44 UTC |
| 8b7049f | paboyle | 19 June 2017, 07:46:07 UTC | Improved detectino of usqcdInfo for plaq/linktr | 19 June 2017, 07:46:07 UTC |
| c850246 | paboyle | 19 June 2017, 00:39:48 UTC | Merge branch 'feature/parallelio' into develop | 19 June 2017, 00:39:48 UTC |
| 1300b0b | paboyle | 19 June 2017, 00:01:48 UTC | Update to enable multiple records per file more consistent with SciDAC. open, close, write records... | 19 June 2017, 00:01:48 UTC |
| e6d984b | paboyle | 17 June 2017, 23:13:22 UTC | ILDG tests | 17 June 2017, 23:13:22 UTC |
| 1d18d95 | paboyle | 17 June 2017, 23:13:03 UTC | Class name return | 17 June 2017, 23:13:03 UTC |
| ae39ec8 | paboyle | 17 June 2017, 23:12:48 UTC | ComplexField defined | 17 June 2017, 23:12:48 UTC |
| b96daf5 | paboyle | 17 June 2017, 23:12:15 UTC | Query tensor structures | 17 June 2017, 23:12:15 UTC |
| 46879e1 | paboyle | 17 June 2017, 23:11:45 UTC | Complex defined in Impl even for gauge. | 17 June 2017, 23:11:45 UTC |
| ae4de94 | paboyle | 17 June 2017, 23:11:23 UTC | SciDAC I/O support | 17 June 2017, 23:11:23 UTC |
| 0ab555b | paboyle | 17 June 2017, 23:11:02 UTC | SciDAC I/O and ILDG improvements | 17 June 2017, 23:11:02 UTC |
| 8e9be9f | paboyle | 17 June 2017, 23:10:42 UTC | Updates for SciDAC IO | 17 June 2017, 23:10:42 UTC |
| d572170 | paboyle | 17 June 2017, 23:10:20 UTC | Update for SciDAC | 17 June 2017, 23:10:20 UTC |
| 81b18f8 | Antonin Portelli | 16 June 2017, 16:59:55 UTC | Merge branch 'feature/scalar_adjointFT' into feature/hadrons # Conflicts: # lib/qcd/action/scalar/ScalarImpl.h | 16 June 2017, 16:59:55 UTC |
| a833f88 | Lanny91 | 16 June 2017, 14:58:47 UTC | Added missing SIMD integer reduction implementation for AVX, AVX-512, SSE4, IMCI | 16 June 2017, 14:58:47 UTC |
| 07b2c1b | Lanny91 | 16 June 2017, 14:04:26 UTC | Placeholder precision change functions to allow Grid to compile with QPX (warning: no actual functionality) | 16 June 2017, 14:04:26 UTC |
| 735cbdb | Lanny91 | 14 June 2017, 09:55:10 UTC | QPX Integer reduction (+ integer reduction test) | 14 June 2017, 09:55:10 UTC |
| 2ad54c5 | Lanny91 | 14 June 2017, 09:53:39 UTC | QPX exchange support | 14 June 2017, 09:53:39 UTC |
| 12ccc73 | paboyle | 14 June 2017, 04:19:17 UTC | Serialisation no compile fix | 14 June 2017, 04:19:17 UTC |
| e7564f8 | paboyle | 13 June 2017, 11:22:50 UTC | Starting a test for reading an ILDG file. | 13 June 2017, 11:22:50 UTC |
| 91199a8 | paboyle | 13 June 2017, 11:21:29 UTC | openmpi is not const safe | 13 June 2017, 11:21:29 UTC |
| 0494fee | paboyle | 13 June 2017, 11:00:23 UTC | Libz dependency | 13 June 2017, 11:00:23 UTC |
| a16b1e1 | paboyle | 13 June 2017, 09:48:43 UTC | gcc 4.9 fix | 13 June 2017, 09:48:43 UTC |
| 769ad57 | paboyle | 11 June 2017, 23:41:21 UTC | Odd new error on G++ 49 on travis | 11 June 2017, 23:41:21 UTC |
| eaac004 | paboyle | 11 June 2017, 23:20:49 UTC | Compile fixes | 11 June 2017, 23:20:49 UTC |
| 56042f0 | paboyle | 11 June 2017, 22:19:20 UTC | New files | 11 June 2017, 22:19:20 UTC |
| 3bfd1f1 | paboyle | 11 June 2017, 22:14:10 UTC | I/O improvements | 11 June 2017, 22:14:10 UTC |
| 70ab598 | Azusa Yamaguchi | 08 June 2017, 21:22:23 UTC | Move gfix into utils | 08 June 2017, 21:22:23 UTC |
| 1d0ca65 | Azusa Yamaguchi | 08 June 2017, 21:21:50 UTC | Move Gfix into utils | 08 June 2017, 21:21:50 UTC |
| 2bc4d0a | Azusa Yamaguchi | 08 June 2017, 21:21:25 UTC | Move code into utils | 08 June 2017, 21:21:25 UTC |
| 2490816 | Antonin Portelli | 08 June 2017, 01:11:02 UTC | Hadrons: rare kaon program removed | 08 June 2017, 01:11:02 UTC |
| 5f55bca | Antonin Portelli | 08 June 2017, 01:10:48 UTC | Hadrons: Quark module renamed MFermion::GaugeProp | 08 June 2017, 01:10:48 UTC |
| f6aa82b | Antonin Portelli | 06 June 2017, 16:46:33 UTC | Merge branch 'develop' into feature/hadrons | 06 June 2017, 16:46:33 UTC |
| 2274969 | Antonin Portelli | 06 June 2017, 16:45:30 UTC | Fixes after merge and point sink module | 06 June 2017, 16:45:30 UTC |
| 0503c02 | Antonin Portelli | 05 June 2017, 21:37:47 UTC | Merge branch 'feature/qed-fvol' into feature/hadrons (non-trivial conflicts on scalar Impl) # Conflicts: # configure.ac # lib/qcd/action/scalar/Scalar.h | 05 June 2017, 21:37:47 UTC |
| 092dcd4 | paboyle | 02 June 2017, 21:50:25 UTC | MPI I/O only if MPI compiled | 02 June 2017, 21:50:25 UTC |
| 4a8c4cc | Guido Cossu | 02 June 2017, 16:02:29 UTC | Test wilson flow, added maxTau for adaptive flow | 02 June 2017, 16:02:29 UTC |
| 9b44189 | Guido Cossu | 02 June 2017, 15:56:00 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 02 June 2017, 15:56:00 UTC |
| 7da4856 | Guido Cossu | 02 June 2017, 15:55:53 UTC | Wilson flow with adaptive steps | 02 June 2017, 15:55:53 UTC |
| aaf1e33 | Guido Cossu | 02 June 2017, 15:32:35 UTC | Adding adaptive integration in the WilsonFlow | 02 June 2017, 15:32:35 UTC |
| 094c3d0 | paboyle | 01 June 2017, 23:38:58 UTC | Improved and RNG's now survive checkpoint | 01 June 2017, 23:38:58 UTC |
| 4b98e52 | Peter Boyle | 01 June 2017, 21:38:18 UTC | Roll over to MPI version of I/O | 01 June 2017, 21:38:18 UTC |
| 1a1f6d5 | Peter Boyle | 01 June 2017, 21:37:26 UTC | Roll over to MPI IO for parallel IO | 01 June 2017, 21:37:26 UTC |
| 2142165 | Peter Boyle | 01 June 2017, 21:36:53 UTC | Big changes improving the code to use MPI IO | 01 June 2017, 21:36:53 UTC |
| 6f687a6 | Peter Boyle | 01 June 2017, 21:36:18 UTC | As local vols increase, use 64 bits for safety | 01 June 2017, 21:36:18 UTC |
| b30754e | paboyle | 30 May 2017, 22:41:28 UTC | Merge branch 'feature/parallelio' of https://github.com/paboyle/Grid into feature/parallelio | 30 May 2017, 22:41:28 UTC |
| 1e429a0 | paboyle | 30 May 2017, 22:41:07 UTC | Added MPI version | 30 May 2017, 22:41:07 UTC |
| d38a4de | paboyle | 30 May 2017, 22:40:39 UTC | Beginning move to MPI IO | 30 May 2017, 22:40:39 UTC |
| ef1b7db | paboyle | 30 May 2017, 22:40:11 UTC | Diff comparison check | 30 May 2017, 22:40:11 UTC |
| 53a9aeb | paboyle | 30 May 2017, 22:39:53 UTC | Cosmetic only | 30 May 2017, 22:39:53 UTC |
| e30fa9f | paboyle | 30 May 2017, 22:39:16 UTC | RankCount; need to clean up ambigious ProcessCount | 30 May 2017, 22:39:16 UTC |
| 58e8d0a | paboyle | 30 May 2017, 22:38:30 UTC | reverse direction lexico mapping | 30 May 2017, 22:38:30 UTC |
| 62cf9cf | paboyle | 30 May 2017, 22:38:02 UTC | Cleaner code | 30 May 2017, 22:38:02 UTC |
| 0fb4588 | paboyle | 30 May 2017, 22:37:02 UTC | Precision safe compile | 30 May 2017, 22:37:02 UTC |
| 725c513 | Peter Boyle | 29 May 2017, 20:47:32 UTC | Better MPI3 benchmarking | 29 May 2017, 20:47:32 UTC |
| d864830 | Antonin Portelli | 29 May 2017, 11:58:08 UTC | Merge branch 'develop' into feature/hadrons | 29 May 2017, 11:58:08 UTC |
| 064315c | Antonin Portelli | 29 May 2017, 11:57:33 UTC | Hadrons: mesons gamma list fix | 29 May 2017, 11:57:33 UTC |
| 7c6cc85 | Guido Cossu | 27 May 2017, 17:03:49 UTC | Updating WilsonFlow test | 27 May 2017, 17:03:49 UTC |
| a6691ef | Guido Cossu | 26 May 2017, 15:43:22 UTC | Merge pull request #110 from Lanny91/feature/hadrons Hadrons: Fermion boundary conditions can now be set in measurement code. | 26 May 2017, 15:43:22 UTC |
| 8e0ced6 | Lanny91 | 26 May 2017, 14:59:15 UTC | Hadrons: Fermion boundary conditions can now be set in measurement code. | 26 May 2017, 14:59:15 UTC |
| 0de3148 | Guido Cossu | 26 May 2017, 13:31:49 UTC | Faster derivative for WilsonGauge | 26 May 2017, 13:31:49 UTC |
| ffb91e5 | Guido Cossu | 26 May 2017, 11:46:02 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 26 May 2017, 11:46:02 UTC |
| f4e8bf2 | Guido Cossu | 26 May 2017, 11:45:59 UTC | Fixing the topological charge. Wilson Flow tested, ok | 26 May 2017, 11:45:59 UTC |
| a74c343 | Antonin Portelli | 25 May 2017, 13:27:49 UTC | Bootstrap script fix | 25 May 2017, 13:27:49 UTC |
| 69470cc | paboyle | 25 May 2017, 12:41:26 UTC | Update to do list | 25 May 2017, 12:41:26 UTC |
| b8b5934 | paboyle | 25 May 2017, 12:32:24 UTC | Attempts to speed up the parallel IO | 25 May 2017, 12:32:24 UTC |
| 75856f2 | Guido Cossu | 25 May 2017, 11:44:56 UTC | Compilation fix in the Tensor_exp | 25 May 2017, 11:44:56 UTC |
| 3c112a7 | Guido Cossu | 25 May 2017, 11:09:00 UTC | Small correction to the general exp definition | 25 May 2017, 11:09:00 UTC |
| ab3596d | Guido Cossu | 25 May 2017, 11:07:47 UTC | Using Cayley-Hamilton form for the exponential of SU(3) matrices | 25 May 2017, 11:07:47 UTC |
| a8c10b1 | paboyle | 25 May 2017, 10:43:33 UTC | Use a global-X x Local-Y chunksize for parallel binary I/O. Gives O(32 x 8 x 18*8*8) chunk size on configuration I/O. At 150KB should be getting close to packet sizes and 4MB filesystem block sizes that are reasonably (!?) performant. We shall see once I move this off my laptop and over to BNL and time it. | 25 May 2017, 10:43:33 UTC |
| 15e801a | Guido Cossu | 19 May 2017, 15:39:36 UTC | Fixing a compilation error for generic SIMD | 19 May 2017, 15:39:36 UTC |
| 0ffc235 | Guido Cossu | 19 May 2017, 09:55:04 UTC | Adding more statistics to the Benchmark_comms. Min and max | 19 May 2017, 09:55:04 UTC |
| 8e19c99 | Guido Cossu | 18 May 2017, 18:07:35 UTC | Adding more statistical info in the Benchmark_comms | 18 May 2017, 18:07:35 UTC |
| a0bc0ad | Guido Cossu | 18 May 2017, 16:48:11 UTC | Reverting change in Bechmark_comms. Keeping 300 iterations | 18 May 2017, 16:48:11 UTC |
| a8fb283 | Guido Cossu | 18 May 2017, 13:45:00 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 18 May 2017, 13:45:00 UTC |
| bc862ce | Guido Cossu | 18 May 2017, 13:44:56 UTC | Fixing an allocation issue in Benchmark_comms | 18 May 2017, 13:44:56 UTC |
| 22f4fee | Antonin Portelli | 17 May 2017, 11:27:13 UTC | Merge branch 'develop' into feature/scalar_adjointFT | 17 May 2017, 11:27:13 UTC |
| 3f858d6 | Antonin Portelli | 17 May 2017, 11:25:14 UTC | Scalar: phi^2 observable | 17 May 2017, 11:25:14 UTC |
| 3267683 | paboyle | 17 May 2017, 10:26:18 UTC | Union workaround for g++ | 17 May 2017, 10:26:18 UTC |
| f46a67f | Azusa Yamaguchi | 17 May 2017, 09:51:01 UTC | No compile issue on clang on mac fixed. Compiler version was clang++-3.9 under mpicxx | 17 May 2017, 09:51:01 UTC |
| f7b8383 | paboyle | 16 May 2017, 13:52:51 UTC | Half precisoin comms mixed prec test | 16 May 2017, 13:52:51 UTC |
