swh:1:snp:f1ea47acc31da44cdcf293f0bcae05f430fc0356

sort by:
Revision Author Date Message Commit Date
d689376 Merge pull request #50 from waterret/develop use sha256 based splittable rng 20 October 2016, 15:40:31 UTC
7af9b87 Cache face tables to improve performance. Extract merge now looking poor. 18 October 2016, 08:51:37 UTC
70f386f switch to use Output Feedback Split mode https://github.com/waterret/RngState-OFS 17 October 2016, 18:20:59 UTC
89cda59 update rng number algo 17 October 2016, 17:31:31 UTC
c39ec3b update license and if guards 17 October 2016, 17:31:31 UTC
8afcc8f fix state size 17 October 2016, 17:31:31 UTC
1abbe2f update rng-state, change output format 17 October 2016, 17:31:31 UTC
4fb37ec fix sprng-sha256 seed with seq 17 October 2016, 17:31:30 UTC
71eaa7c use sha256 based splittable rng 17 October 2016, 17:31:30 UTC
811ca45 GNU clang hack for AVX512 since there are missing reduce intrinsics in Clang 3.9 and GCC-6 AVX512 support 17 October 2016, 15:23:21 UTC
bc1a4d4 Faster integer handling avoid push_back 17 October 2016, 15:16:44 UTC
c8079e6 Time the face gateher in x-dir more carefully 13 October 2016, 21:28:50 UTC
8b0d171 32bit issue on the KNL code variant where byte offsets were stored 12 October 2016, 16:49:32 UTC
1f293b7 Merge branch 'feature/knl-stats' into develop 12 October 2016, 12:47:58 UTC
8bbd9eb Reversing changes to Stencil class 12 October 2016, 12:47:20 UTC
6472b43 __rdpmc needed for gcc, clang++ 12 October 2016, 11:29:08 UTC
bd205a3 Fixing for non x86 and non KNL 12 October 2016, 11:09:15 UTC
496beff Fix non-KNL build 12 October 2016, 11:06:08 UTC
9b63e97 align not absolutely required and confuses clang++ 12 October 2016, 10:51:21 UTC
81f2aea KNL streaming stores, and KNL performance coutners 12 October 2016, 10:45:22 UTC
2d4a45c Typecast pointer 12 October 2016, 08:14:15 UTC
0f182f0 Drop macos with gcc 11 October 2016, 21:29:06 UTC
7240d73 Parallelise the x faces; fix the segv on KNL with comms 11 October 2016, 21:21:07 UTC
42cd148 Base pointer for comms buffer under AVX512 assembly 11 October 2016, 15:06:06 UTC
611b5d7 Fix for AVX+FMA3 compilation 10 October 2016, 14:26:17 UTC
b56c9ff Fix for AVXFMA 10 October 2016, 13:43:37 UTC
70c32fa Merge branch 'develop' of github.com:paboyle/Grid into develop 09 October 2016, 11:55:46 UTC
77c8a94 AVXFMA4 flag fix for Intel Compiler 09 October 2016, 11:55:12 UTC
2e453df Added some instrumentation to benchmark the force computation 06 October 2016, 16:52:45 UTC
4089984 Timing hooks 06 October 2016, 08:25:12 UTC
9843984 configure portability fix 05 October 2016, 13:57:20 UTC
c78bbd0 Fix ASM compilation 04 October 2016, 14:37:32 UTC
7ea4b95 hopefully more portable configure output 27 September 2016, 10:54:37 UTC
536e2ff *.inc removed: please don't commit these files either! 27 September 2016, 10:54:03 UTC
798ff34 configure removed: please don't commit configure! 27 September 2016, 10:29:31 UTC
04a437c Minor modification to the filelist script 23 September 2016, 10:12:45 UTC
5c190a1 Merge branch 'develop' into feature/hirep 23 September 2016, 10:06:06 UTC
15d8f5c Small change to the configure.ac to include the canonical names 23 September 2016, 10:05:36 UTC
c4ac6e7 Consolidating HMC interface Uniformed interface for standard action in fundamental rep and Hirep 23 September 2016, 09:47:42 UTC
510e340 Debugged last commit for the Two index representation 22 September 2016, 21:16:21 UTC
6ffadca Restored number of colours to 3 22 September 2016, 13:22:54 UTC
b6597b7 Added support for the Two index Symmetric and Antisymmetric representations Tested for HMC convergence: OK Added also a test file showing an example for mixed representations 22 September 2016, 13:17:37 UTC
d257318 build system: FFTW fix 20 September 2016, 11:30:24 UTC
65ca174 gitignore update 20 September 2016, 10:25:06 UTC
0724f7a QPX single precision implementation 19 September 2016, 17:09:12 UTC
2e74520 removed libtool use (BG/Q compatibility) 16 September 2016, 14:25:49 UTC
6dd75ad Merge branch 'develop' of github.com:paboyle/Grid into feature/bgq 16 September 2016, 14:07:54 UTC
fda408e Added first lines for supporting Two Index representations 13 September 2016, 09:43:30 UTC
b9c8031 Merge branch 'develop' into feature/hirep 13 September 2016, 09:01:51 UTC
5df5d52 Fix for the Intel compiler 12 September 2016, 16:17:20 UTC
f76f281 Cleaning files after fix 09 September 2016, 10:34:25 UTC
aa20cc8 Fixing compilation error with AVX512 flag 09 September 2016, 09:58:52 UTC
0fd179f Merge branch 'develop' into feature/hirep 01 September 2016, 11:59:53 UTC
f45ef8d Minor modification in ActionBase.h 01 September 2016, 10:46:46 UTC
fd56147 Merge branch 'develop' into feature/hirep 30 August 2016, 17:21:36 UTC
005dcc5 Reset travis 30 August 2016, 13:44:10 UTC
655c893 Another test on travis 30 August 2016, 13:38:42 UTC
843f578 Again travis test separating single and double 30 August 2016, 13:29:09 UTC
8986c9f Single and double precision travis matrix 30 August 2016, 13:25:24 UTC
c80a1d4 Retest original version of travis yaml 30 August 2016, 13:05:05 UTC
ae57032 Separate single and double builds in travis 30 August 2016, 13:00:34 UTC
f754687 Another error on travis 30 August 2016, 12:56:23 UTC
5acd856 Correction of error in travis 30 August 2016, 12:49:49 UTC
b0d3e4b Separating travis builds 30 August 2016, 12:44:07 UTC
b512ccb HMC for Adjoint fermions works Accepts and reproduces known results Check initial instability of inverters when starting from hot configurations 30 August 2016, 10:31:25 UTC
8c89391 FFTW unresolved fixed when no fftw3.h 24 August 2016, 15:41:47 UTC
bfac519 tidy up 24 August 2016, 15:38:36 UTC
a782ca3 Merge branch 'feature/fft-flop-count' into develop 24 August 2016, 14:06:17 UTC
7446910 Printing 24 August 2016, 14:05:56 UTC
ff6da36 FFT double and single precision gives good performance now in multithreaded code. 24 August 2016, 14:05:00 UTC
4d11a6f first commit for QPX intrinsics 23 August 2016, 13:41:44 UTC
88be3b3 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 22 August 2016, 17:29:36 UTC
8a02824 Merge branch 'feature/FFT' into develop 22 August 2016, 15:25:04 UTC
356e794 fftw can be switched off 22 August 2016, 15:24:49 UTC
73ce476 Include fftw headers 22 August 2016, 15:24:21 UTC
29c4ef4 Adding a test for libfftw3 22 August 2016, 15:21:01 UTC
e423a09 FFT improved and test_FFT passing under MPI 8 processes, 8^4 for LatticeComplexD and LatticeSpinMatrixD 18 August 2016, 01:23:21 UTC
17097a9 FFTW test ran over 4 mpi processes. 17 August 2016, 00:33:55 UTC
94a6373 Merge branch 'feature/eigen-cleanup' into develop 15 August 2016, 22:58:34 UTC
4ab7dbf Instantiate 15 August 2016, 22:00:40 UTC
90e7079 Feature for z-Mobius prep 15 August 2016, 21:31:29 UTC
9c2e8d5 Nc=3 just to let all the test pass in Travis 09 August 2016, 14:46:57 UTC
147e202 Added unit tests on the representation transformations Status: Passing all tests 08 August 2016, 15:54:22 UTC
573b8c6 build system: -O3 is not overriden by env CXXFLAGS 06 August 2016, 00:26:24 UTC
15218ec more Travis MPI fix 05 August 2016, 23:49:14 UTC
ec68e08 Travis MPI fix 05 August 2016, 23:36:05 UTC
fc25d22 fftw download 05 August 2016, 23:28:52 UTC
8dc2cfc Adding fftw header pulling 05 August 2016, 23:28:28 UTC
836f937 first try at including MPI tests in Travis 05 August 2016, 12:41:52 UTC
5a68715 Richards sweep test 05 August 2016, 09:51:57 UTC
32bc7a6 MPI back out of change that hangs AVX2 for clang, gcc needs the -mfma flag. 05 August 2016, 09:36:00 UTC
b65e72e Merge pull request #43 from rprollins/bench/output-format Benchmark_dwf_sweep and Benchmark_zmm output formats 04 August 2016, 15:47:01 UTC
d1aaff6 README update 04 August 2016, 15:27:02 UTC
93d29bb build system improvements after discussion with Peter 04 August 2016, 15:19:59 UTC
3b376ed build system: error if MPI not found 03 August 2016, 14:23:38 UTC
d5c1f61 gitignore update 03 August 2016, 14:14:33 UTC
2edc242 untracking ltmain.sh 03 August 2016, 14:12:44 UTC
6292837 build system: local Grid link flag moved to configure.ac 03 August 2016, 14:07:42 UTC
6adb66d build system: finer management of GMP/MPFR dependence 03 August 2016, 14:06:45 UTC
5be92bb link fix in README 03 August 2016, 11:40:56 UTC
back to top