d4290a7 | Antonin Portelli | 17 June 2021, 10:57:02 UTC | finer timers in Benchmark_IO | 17 June 2021, 10:57:02 UTC |
92def28 | Peter Boyle | 06 June 2021, 08:52:05 UTC | Update README.md | 06 June 2021, 08:52:05 UTC |
ca10bfa | Antonin Portelli | 04 June 2021, 10:12:22 UTC | removing Travis CI constantly failing due to overtime (no way we can compile Grid on free time anymore) | 04 June 2021, 10:12:22 UTC |
0e27e38 | Peter Boyle | 03 June 2021, 04:24:19 UTC | Remove synch | 03 June 2021, 04:24:19 UTC |
8cfc734 | u61464 | 05 May 2021, 21:17:18 UTC | staggered hand unroll read coalesce | 05 May 2021, 21:17:18 UTC |
15ae317 | u61464 | 04 May 2021, 15:40:38 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 04 May 2021, 15:40:38 UTC |
834f536 | u61464 | 04 May 2021, 15:40:18 UTC | Fastest option on SyCL is now std::complex | 04 May 2021, 15:40:18 UTC |
c332d9f | Peter Boyle | 27 April 2021, 18:10:49 UTC | Merge pull request #356 from felixerben/bugfix/stoutSmearing Jamie's fix | 27 April 2021, 18:10:49 UTC |
cf2923d | Felix Erben | 27 April 2021, 15:53:37 UTC | Jamie's fix | 27 April 2021, 15:53:37 UTC |
0e4413d | Peter Boyle | 27 April 2021, 12:01:55 UTC | Merge pull request #355 from felixerben/bugfix/stoutSmearing bugfix 3D stout smearing | 27 April 2021, 12:01:55 UTC |
009ccd5 | Felix Erben | 26 April 2021, 09:36:33 UTC | bugfix 3D stout smearing | 26 April 2021, 09:36:33 UTC |
8cd4263 | Peter Boyle | 26 April 2021, 02:20:37 UTC | Tests compile | 26 April 2021, 02:20:37 UTC |
d45c868 | Peter Boyle | 25 April 2021, 14:53:34 UTC | Change interface | 25 April 2021, 14:53:34 UTC |
955a811 | Peter Boyle | 25 April 2021, 14:36:38 UTC | Expose label only to reduce number of parameters | 25 April 2021, 14:36:38 UTC |
dbe210d | Peter Boyle | 25 April 2021, 14:25:59 UTC | Open the ens_id | 25 April 2021, 14:25:59 UTC |
86e1174 | Peter Boyle | 20 April 2021, 14:19:11 UTC | set twists | 20 April 2021, 14:19:11 UTC |
980e721 | Peter Boyle | 13 April 2021, 13:33:01 UTC | Update MetaData.h | 13 April 2021, 13:33:01 UTC |
e2a0142 | Peter Boyle | 06 April 2021, 14:49:00 UTC | Merge pull request #348 from AndrewYongZhenNing/develop Conserved Tadpole Implementation for Shamir Action Only | 06 April 2021, 14:49:00 UTC |
895244e | Andrew Zhen Ning Yong | 06 April 2021, 12:46:33 UTC | Merge with upstream; implemented conserved tadpole for Shamir action. | 06 April 2021, 12:46:33 UTC |
addeb62 | Andrew Zhen Ning Yong | 06 April 2021, 12:45:37 UTC | Implemented tadpole operator for Shamir action. | 06 April 2021, 12:45:37 UTC |
a7fb25a | Peter Boyle | 29 March 2021, 19:44:14 UTC | Make Cshift fields static to avoid repeated reallocaate overhead | 29 March 2021, 19:44:14 UTC |
e947992 | Peter Boyle | 29 March 2021, 18:04:06 UTC | Improved force terms | 29 March 2021, 18:04:06 UTC |
bb89a82 | Peter Boyle | 29 March 2021, 18:01:15 UTC | Staggered coalseced read | 29 March 2021, 18:01:15 UTC |
8bdadba | Peter Boyle | 18 March 2021, 19:41:14 UTC | Cold start | 18 March 2021, 19:41:14 UTC |
15c50a7 | Peter Boyle | 18 March 2021, 19:40:42 UTC | Explicit instantiate the template function | 18 March 2021, 19:40:42 UTC |
49b0af2 | Peter Boyle | 18 March 2021, 13:10:02 UTC | Update of tests to compile with the sRNG addition. Audited the code conventions (again) with the CPS momentum denominator and added anti periodic in time to the Test_mobius_force.cc and tested the Test_dwf_gpforce. Promoted thesee to test full HMC hamiltonian, tr P^2/2 + phidag MdagM phi with the same pdot and Udot as audited in the Integrator.h etc... With full comments and sources for factors. | 18 March 2021, 13:10:02 UTC |
9c2b372 | Peter Boyle | 18 March 2021, 10:24:11 UTC | sRNG parameter added | 18 March 2021, 10:24:11 UTC |
3c67d62 | Peter Boyle | 12 March 2021, 14:36:55 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 12 March 2021, 14:36:55 UTC |
51f5065 | Peter Boyle | 12 March 2021, 14:33:04 UTC | Read out the local ID once, and store | 12 March 2021, 14:33:04 UTC |
226be84 | Peter Boyle | 12 March 2021, 14:31:50 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 12 March 2021, 14:31:50 UTC |
001814b | Peter Boyle | 12 March 2021, 14:31:17 UTC | updated to do list. Start adding DDHMC work items | 12 March 2021, 14:31:17 UTC |
db3ac67 | Peter Boyle | 12 March 2021, 13:55:07 UTC | Update thread issue | 12 March 2021, 13:55:07 UTC |
da91a88 | Peter Boyle | 11 March 2021, 22:54:53 UTC | NVCC versions found buggy added as guard | 11 March 2021, 22:54:53 UTC |
a71e675 | Peter Boyle | 11 March 2021, 21:43:06 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 11 March 2021, 21:43:06 UTC |
cd5891e | Peter Boyle | 11 March 2021, 21:34:28 UTC | Test that fails on Cuda 11.0 | 11 March 2021, 21:34:28 UTC |
5bb7336 | Peter Boyle | 11 March 2021, 21:29:07 UTC | Merge pull request #347 from pjgeorg/fix-autotools-avx512 Fix inconsistent SIMD option AVX512 Thanks | 11 March 2021, 21:29:07 UTC |
ce1fc1f | Peter Boyle | 11 March 2021, 21:20:53 UTC | Possible fallback plan for Fionn's compiler bbug in nvcc | 11 March 2021, 21:20:53 UTC |
82402c6 | Peter Georg | 11 March 2021, 12:08:40 UTC | Add simd option SKL for ICC | 11 March 2021, 12:08:40 UTC |
d9c4afe | Peter Georg | 11 March 2021, 11:58:49 UTC | Fix inconsistent configure option AVX512 Before this change AVX512 enabled different instruction sets depending on the compiler: For Intel C++ Compiler Classic (ICC): AVX512F, AVX512CD, AVX512DQ, AVX512BW, AVX512VL i.e. Intel Xeon Skylake and newer For Intel ICX, gcc, clang: AVX512F, AVX512CD, AVX512ER, AVX512PF i.e. Intel Xeon Phi x200/x205 (KNL/KNM) With this commit AVX512 now only enables the common instruction sets supported by all CPUs supporting any AVX-512 instructions set: AVX512F and AVX512CD (called COMMON-AVX512 by icc) | 11 March 2021, 11:58:49 UTC |
f786ff8 | Peter Boyle | 10 March 2021, 19:32:06 UTC | Extend test from Fionn, fails on A100 apparently | 10 March 2021, 19:32:06 UTC |
a651cae | u61464 | 10 March 2021, 14:23:51 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 10 March 2021, 14:23:51 UTC |
0e21adb | u61464 | 10 March 2021, 13:40:51 UTC | Gives 200GF/s on SyCL/DG1 8^4, doesn't uglify develop for other platforms too badly. Easy to revert to clean more C++ stylistic code. Theres a SYCL_HACK macro I will clean up later once dpcpp evolves a central nervous systems. | 10 March 2021, 13:40:51 UTC |
58bf9b9 | Peter Boyle | 10 March 2021, 01:45:22 UTC | Clean up test | 10 March 2021, 01:45:22 UTC |
2146eeb | Peter Boyle | 09 March 2021, 03:31:46 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 09 March 2021, 03:31:46 UTC |
6a429ee | Peter Boyle | 09 March 2021, 03:31:10 UTC | 2d loop hits Nvidia 16bit limit on large local vols | 09 March 2021, 03:31:10 UTC |
4d1ea15 | Peter Boyle | 09 March 2021, 03:29:37 UTC | More verbosity. The 16bit limit on Grid.y, Grid.z is annoying | 09 March 2021, 03:29:37 UTC |
a76cb00 | Peter Boyle | 08 March 2021, 18:37:57 UTC | Update Tensor_exp.h | 08 March 2021, 18:37:57 UTC |
a960436 | Peter Boyle | 05 March 2021, 18:17:19 UTC | Merge pull request #336 from lehner/feature/gpt Make ShmDims configurable; adjust GRID_MAX_SIMD to allow for 128 byte width on GPUs | 05 March 2021, 18:17:19 UTC |
d706502 | Peter Boyle | 05 March 2021, 17:47:07 UTC | Merge pull request #332 from mmphys/feature/mres_schur Optional changes to Test_cayley_mres e.g. Schur solver | 05 March 2021, 17:47:07 UTC |
89d299c | Peter Boyle | 05 March 2021, 17:46:33 UTC | Merge pull request #333 from mmphys/bugfix/LatTransfer Fix convertType for GPU in Lattice_transfer.h | 05 March 2021, 17:46:33 UTC |
e34eda6 | Peter Boyle | 05 March 2021, 17:45:44 UTC | Merge pull request #344 from felixerben/feature/XiToSigma Feature/xi to sigma | 05 March 2021, 17:45:44 UTC |
b24181a | Christoph Lehner | 05 March 2021, 15:56:58 UTC | Update Coordinate.h Revert GRID_MAX_SIMD change | 05 March 2021, 15:56:58 UTC |
aa173e2 | Peter Boyle | 05 March 2021, 15:25:33 UTC | Update README.md | 05 March 2021, 15:25:33 UTC |
7a19432 | Felix Erben | 05 March 2021, 10:57:09 UTC | whitespace | 05 March 2021, 10:57:09 UTC |
9b15704 | Felix Erben | 05 March 2021, 10:42:32 UTC | tested and consitent | 05 March 2021, 10:42:32 UTC |
017f955 | Michael Marshall | 04 March 2021, 20:42:02 UTC | Merge branch 'develop' into feature/mres_schur * develop: Pass serial RNG around Sycl happier | 04 March 2021, 20:42:02 UTC |
f252d69 | Michael Marshall | 04 March 2021, 20:41:30 UTC | Merge branch 'develop' into bugfix/LatTransfer * develop: Pass serial RNG around Sycl happier | 04 March 2021, 20:41:30 UTC |
3b06e46 | Felix Erben | 04 March 2021, 20:06:16 UTC | Merge branch 'develop' into feature/XiToSigma | 04 March 2021, 20:06:16 UTC |
d4b4de8 | Felix Erben | 04 March 2021, 20:01:24 UTC | changes | 04 March 2021, 20:01:24 UTC |
c90beee | Peter Boyle | 03 March 2021, 22:50:29 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 03 March 2021, 22:50:29 UTC |
1eea9d7 | Peter Boyle | 03 March 2021, 22:50:01 UTC | Pass serial RNG around | 03 March 2021, 22:50:01 UTC |
679d1d2 | u61464 | 03 March 2021, 19:21:43 UTC | Sycl happier | 03 March 2021, 19:21:43 UTC |
b2b5e0b | Michael Marshall | 03 March 2021, 16:15:12 UTC | Merge branch 'develop' into feature/mres_schur * develop: Hand unrolled to use optimised code paths on GPU for coalesced reads in Wilson case. Other cases to do. This now includes comms code path. Better SIMD usage/coalescence | 03 March 2021, 16:15:12 UTC |
03e5472 | Michael Marshall | 03 March 2021, 16:13:23 UTC | Merge branch 'develop' into bugfix/LatTransfer * develop: Hand unrolled to use optimised code paths on GPU for coalesced reads in Wilson case. Other cases to do. This now includes comms code path. | 03 March 2021, 16:13:23 UTC |
442336b | Peter Boyle | 02 March 2021, 13:50:51 UTC | Hand unrolled to use optimised code paths on GPU for coalesced reads in Wilson case. Other cases to do. This now includes comms code path. | 02 March 2021, 13:50:51 UTC |
9c9566b | Christoph Lehner | 01 March 2021, 11:33:51 UTC | Merge pull request #23 from paboyle/develop Sync | 01 March 2021, 11:33:51 UTC |
1059a81 | Michael Marshall | 27 February 2021, 00:21:36 UTC | Merge branch 'develop' into bugfix/LatTransfer * develop: Better SIMD usage/coalescence | 27 February 2021, 00:21:36 UTC |
2e61556 | Peter Boyle | 26 February 2021, 16:52:20 UTC | Merge branch 'develop' of https://github.com/paboyle/Grid into develop | 26 February 2021, 16:52:20 UTC |
f9b1f24 | Peter Boyle | 26 February 2021, 16:51:41 UTC | Better SIMD usage/coalescence | 26 February 2021, 16:51:41 UTC |
69f4146 | Michael Marshall | 25 February 2021, 09:19:17 UTC | Merge branch 'develop' into bugfix/LatTransfer * develop: (26 commits) Added the ability to apply a custom "filter" to the conjugate momentum in the Integrator classes, applied both after refresh and after applying the forces Added a conjugate momentum "filter" that applies a phase to each site. With sites set to 1.0 or 0.0 this acts as a mask and enables, for example, the freezing of inactive gauge links in DDHMC Added tests/forces/Test_momentum_filter demonstrating the use of the filter to freeze boundary links Correct misleading ac help string Enable performance counting in WilsonFermion like in others changed back A2AUtils warning changed if and accelerator_for - no runtime errors any more Mac OS (Darwin) sed -i flag for in-place editing differs from posix / gnu Seems the intention with AutoConf produced Grid/Config.h was to use sed to translate standard PACKAGE_ #defines into GRID_ however due to missing '' after -i this hasn't been working. Perhaps it is too late to fix this, since we don't know who/what is relying on this downstream? ... but if they are, and AutoConf is being used, then likely these #defines have just been redefined anyway. Seems reasonable to redefine PACKAGE and VERSION as well, as none of these macros are used throughout Grid or Hadrons. Fixed compile issues with maxLocalNorm2 for non-scalar lattices maxLocalNorm2 test now reuses the random field MADWF 5d source option for hadrons - look at Grid of source Abort on GPU error maxLocalNorm2() change back benchmark_ITT prettify Flop cout matches DiRAC-ITT-2020 revert changes merge develop fixes weird bug in 2pt function... revert changes final version, tested on CPU and GPU bugfix ... | 25 February 2021, 09:19:17 UTC |
d620b30 | Michael Marshall | 24 February 2021, 18:07:27 UTC | Merge branch 'develop' into feature/mres_schur * develop: (26 commits) Added the ability to apply a custom "filter" to the conjugate momentum in the Integrator classes, applied both after refresh and after applying the forces Added a conjugate momentum "filter" that applies a phase to each site. With sites set to 1.0 or 0.0 this acts as a mask and enables, for example, the freezing of inactive gauge links in DDHMC Added tests/forces/Test_momentum_filter demonstrating the use of the filter to freeze boundary links Correct misleading ac help string Enable performance counting in WilsonFermion like in others changed back A2AUtils warning changed if and accelerator_for - no runtime errors any more Mac OS (Darwin) sed -i flag for in-place editing differs from posix / gnu Seems the intention with AutoConf produced Grid/Config.h was to use sed to translate standard PACKAGE_ #defines into GRID_ however due to missing '' after -i this hasn't been working. Perhaps it is too late to fix this, since we don't know who/what is relying on this downstream? ... but if they are, and AutoConf is being used, then likely these #defines have just been redefined anyway. Seems reasonable to redefine PACKAGE and VERSION as well, as none of these macros are used throughout Grid or Hadrons. Fixed compile issues with maxLocalNorm2 for non-scalar lattices maxLocalNorm2 test now reuses the random field MADWF 5d source option for hadrons - look at Grid of source Abort on GPU error maxLocalNorm2() change back benchmark_ITT prettify Flop cout matches DiRAC-ITT-2020 revert changes merge develop fixes weird bug in 2pt function... revert changes final version, tested on CPU and GPU bugfix ... | 24 February 2021, 18:07:27 UTC |
157fd14 | Peter Boyle | 24 February 2021, 16:29:52 UTC | Merge pull request #342 from paboyle/feature/link-update-mask Feature/link update mask | 24 February 2021, 16:29:52 UTC |
c791cb2 | Christopher Kelly | 23 February 2021, 16:51:54 UTC | Merge branch 'develop' into feature/link-update-mask | 23 February 2021, 16:51:54 UTC |
d5ab571 | Christopher Kelly | 23 February 2021, 16:49:56 UTC | Added the ability to apply a custom "filter" to the conjugate momentum in the Integrator classes, applied both after refresh and after applying the forces Added a conjugate momentum "filter" that applies a phase to each site. With sites set to 1.0 or 0.0 this acts as a mask and enables, for example, the freezing of inactive gauge links in DDHMC Added tests/forces/Test_momentum_filter demonstrating the use of the filter to freeze boundary links | 23 February 2021, 16:49:56 UTC |
0ed800f | Felix Erben | 23 February 2021, 14:54:46 UTC | merge develop | 23 February 2021, 14:54:46 UTC |
0a32183 | Peter Boyle | 23 February 2021, 14:30:16 UTC | Merge pull request #335 from felixerben/gpu/baryons Gpu/baryons | 23 February 2021, 14:30:16 UTC |
2cacfbd | Peter Boyle | 22 February 2021, 14:28:50 UTC | Merge pull request #341 from DanielRichtmann/fix/minor-things Minor fixes | 22 February 2021, 14:28:50 UTC |
c073e62 | Daniel Richtmann | 22 February 2021, 14:17:07 UTC | Correct misleading ac help string | 22 February 2021, 14:25:44 UTC |
e3d019b | Daniel Richtmann | 22 February 2021, 13:56:52 UTC | Enable performance counting in WilsonFermion like in others | 22 February 2021, 14:25:40 UTC |
7ae030f | Felix Erben | 18 February 2021, 13:24:50 UTC | changed back A2AUtils warning | 18 February 2021, 13:24:50 UTC |
86b58d5 | Felix Erben | 18 February 2021, 12:04:32 UTC | changed if and accelerator_for - no runtime errors any more | 18 February 2021, 12:04:32 UTC |
26e8b9f | Peter Boyle | 17 February 2021, 16:56:21 UTC | Merge pull request #340 from mmphys/bugfix/config Mac OS (Darwin) sed -i flag for in-place editing differs from posix / gnu | 17 February 2021, 16:56:21 UTC |
35114c9 | Michael Marshall | 17 February 2021, 13:24:15 UTC | Mac OS (Darwin) sed -i flag for in-place editing differs from posix / gnu | 17 February 2021, 13:24:15 UTC |
dfd28a8 | Peter Boyle | 15 February 2021, 18:53:26 UTC | Merge pull request #339 from mmphys/bugfix/config Optional rename PACKAGE_ to GRID_ in Grid/Config.h | 15 February 2021, 18:53:26 UTC |
a503332 | Michael Marshall | 14 February 2021, 21:27:54 UTC | Seems the intention with AutoConf produced Grid/Config.h was to use sed to translate standard PACKAGE_ #defines into GRID_ however due to missing '' after -i this hasn't been working. Perhaps it is too late to fix this, since we don't know who/what is relying on this downstream? ... but if they are, and AutoConf is being used, then likely these #defines have just been redefined anyway. Seems reasonable to redefine PACKAGE and VERSION as well, as none of these macros are used throughout Grid or Hadrons. | 14 February 2021, 21:27:54 UTC |
1ac13ec | Peter Boyle | 08 February 2021, 17:08:11 UTC | Merge pull request #338 from paboyle/bugfix/maxnorm2 Fixed compile issues with maxLocalNorm2 for non-scalar lattices | 08 February 2021, 17:08:11 UTC |
55de69a | Christopher Kelly | 08 February 2021, 17:03:16 UTC | Fixed compile issues with maxLocalNorm2 for non-scalar lattices maxLocalNorm2 test now reuses the random field | 08 February 2021, 17:03:16 UTC |
eda9ab4 | Peter Boyle | 08 February 2021, 15:47:22 UTC | MADWF 5d source option for hadrons - look at Grid of source Abort on GPU error | 08 February 2021, 15:47:22 UTC |
cd99edc | Peter Boyle | 04 February 2021, 23:25:49 UTC | maxLocalNorm2() | 04 February 2021, 23:25:49 UTC |
4705aa5 | Christoph Lehner | 04 February 2021, 13:25:55 UTC | Allow user to configure ShmDims via environment variables | 04 February 2021, 13:25:55 UTC |
3215d88 | Michael Marshall | 03 February 2021, 15:17:03 UTC | Simplify syntax with Grid::EnableIf post code review. Updated EnableIf so that ReturnType defaults to void in same way as std::enable_if see https://en.cppreference.com/w/cpp/types/enable_if | 03 February 2021, 15:17:03 UTC |
9b9a53f | Felix Erben | 02 February 2021, 13:06:43 UTC | ... | 02 February 2021, 13:06:43 UTC |
019ffe1 | Christoph Lehner | 02 February 2021, 10:32:23 UTC | Allow for GPU vector width beyond 64 | 02 February 2021, 10:32:23 UTC |
bc496dd | Felix Erben | 28 January 2021, 14:29:56 UTC | change back benchmark_ITT | 28 January 2021, 14:29:56 UTC |
a673b6a | Felix Erben | 28 January 2021, 14:15:09 UTC | prettify | 28 January 2021, 14:15:09 UTC |
1bf2e4d | Felix Erben | 27 January 2021, 21:17:37 UTC | Merge branch 'develop' into gpu/baryons | 27 January 2021, 21:17:37 UTC |
96dd7a8 | Peter Boyle | 16 November 2020, 16:15:34 UTC | Flop cout matches DiRAC-ITT-2020 | 27 January 2021, 21:14:52 UTC |
7905afa | Felix Erben | 19 January 2021, 12:32:48 UTC | revert changes | 27 January 2021, 21:14:52 UTC |
712bb40 | Felix Erben | 15 December 2020, 16:33:29 UTC | merge develop | 27 January 2021, 21:14:52 UTC |
81d88d9 | Felix Erben | 27 January 2021, 21:09:51 UTC | fixes | 27 January 2021, 21:09:51 UTC |