moebiusband/MD-Bench
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
333 Commits 8 Branches 0 Tags 3.4 MiB
da3b1dd53f
Commit Graph

56 Commits

Author SHA1 Message Date
Rafael Ravedutti
da3b1dd53f Add extended parameter option --param 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2023-11-21 15:27:11 +01:00
Rafael Ravedutti
2f13291817 Change function get_num_threads to get_cuda_num_threads 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2023-11-21 14:40:19 +01:00
Rafael Ravedutti
a460fffa19 Fix PBC case 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2023-10-10 12:53:43 +02:00
Rafael Ravedutti
8253b31ee0 Include masked out interactions from remainder in atoms_outside_cutoff 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2023-04-09 03:44:21 +02:00
Rafael Ravedutti
b6982d56f5 Fix atom sorting 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2023-04-09 01:19:12 +02:00
Rafael Ravedutti
039de0be99 Fix stubbed versions and debug messages 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2023-03-30 03:49:57 +02:00
JanLJL
d0277765c3 changed double constants to MD_FLOAT constants 2023-02-13 14:13:53 +01:00
JanLJL
98583cdade Merge branch 'master' of github.com:RRZE-HPC/MD-Bench 2023-02-09 17:33:42 +01:00
JanLJL
3b076cdb49 changed double consts to MD_FLOAT 2023-02-09 17:33:07 +01:00
Rafael Ravedutti
122a23e2b8 Fix compilation error when not using explicit SIMD version 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2023-02-07 23:53:32 +01:00
Yannick Paschke
c61cf9a0ac Move likwid marker calls into OpenMP parallel region 2023-01-22 15:33:05 +01:00
Rafael Ravedutti
15d43dcce5 Explicitly set half_neigh to zero on stubbed versions 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-12-14 17:21:09 +01:00
Rafael Ravedutti
cd1fbfb3c8 Reorganize SIMD files and split AVX and AVX2 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-11-15 00:55:46 +01:00
Rafael Ravedutti
c70ebce4c1 Integrate GROMACS GPU implementation into master branch 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-11-08 18:33:23 +01:00
Rafael Ravedutti
493915fe95 Fix code for AVX and remove warnings 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-11-08 15:30:37 +01:00
Rafael Ravedutti
437b380229 Adjust NVCC flags 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-11-07 20:37:01 +01:00
Jan Eitzinger
3d0f4b97ee Switch copyright header in source files. 2022-09-05 10:39:42 +02:00
Rafael Ravedutti
28d3946072 Move common modules to common directory 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-17 17:56:31 +02:00
Rafael Ravedutti
47db9e86b0 Introduce common directory 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-17 17:20:57 +02:00
Rafael Ravedutti
911ba63336 Adjust ISA options and improve output 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-16 18:36:47 +02:00
Rafael Ravedutti
0caeea0494 Rename cuda.c to device.c 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-12 18:17:07 +02:00
Rafael Ravedutti
90609a2b5f Adjust file structure for CUDA 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-12 18:12:29 +02:00
Rafael Ravedutti
939197a785 Create separate structs DeviceAtom and DeviceNeighbor with device pointers 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-12 17:28:06 +02:00
Rafael Ravedutti
065b596074 Initial refactoring of CUDA code 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-12 04:19:38 +02:00
Rafael Ravedutti
959ff65126 Fix macro condition 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-12 01:29:40 +02:00
Rafael Ravedutti
87d006d418 Fix GPU version 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-11 16:42:41 +02:00
Rafael Ravedutti
3d95ec4b0a Small fixes 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-09 19:19:48 +02:00
Rafael Ravedutti
c18124b066 Integrate LAMMPS CUDA versions into master branch 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-08-09 18:53:53 +02:00
Rafael Ravedutti
eb77e1a3bd Fix DEM setup 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-19 04:13:06 +02:00
Rafael Ravedutti
2e77f6207b Avoid errors when compiling for AVX2 due to SIMD LJ implementation 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-19 02:30:26 +02:00
Rafael Ravedutti
577955dfb7 Apply first changes to DEM kernel 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-13 02:34:33 +02:00
Rafael Ravedutti
99237241fb Include domain box on DEM input file 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-08 23:15:30 +02:00
Rafael Ravedutti
814f561993 Allow PBC in just some directions 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-08 02:30:03 +02:00
Rafael Ravedutti
32836eebcb Setup first DEM example with input file from lecture 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-07 02:11:50 +02:00
Rafael Ravedutti
9ffc09f497 Add DEM kernel to parameter options 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-07 00:47:38 +02:00
Rafael Ravedutti
79483a446e Adjust code with DEM to be compilable 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-06 01:07:39 +02:00
Rafael Ravedutti
bb599c9ea8 Add first version of DEM kernel 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-07-05 15:33:31 +02:00
Rafael Ravedutti
e4d7faf91b Adjust cutforce and atom positions in stubbed version 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-05-14 01:02:08 +02:00
Rafael Ravedutti
bbdcaf2983 New stubbed version 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-05-14 00:55:33 +02:00
Rafael Ravedutti
14838389ff Fix stubbed variant for LAMMPS algorithm 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-04-30 04:08:18 +02:00
Rafael Ravedutti
ab2eb1ff50 Write LAMMPS kernel with SIMD intrinsics and implement AVX512 with double-precision functions 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-04-05 02:57:23 +02:00
Jan Eitzinger
5585ebcf42 Add ONEAPI config. Remove omp simd for full neigh. 2022-04-01 15:57:54 +02:00
Rafael Ravedutti
2a555a7deb Add simd reduction pragma to vectorize innermost loop on half-neighbor variant 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-03-21 17:02:09 +01:00
Rafael Ravedutti
719330807b Change data layout for force arrays according to position 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-03-18 01:40:51 +01:00
Rafael Ravedutti
e7737e9151 Refactor half neighbor lists code 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-03-18 01:28:11 +01:00
Rafael Ravedutti
5df544637f Fix force calculation time in LAMMPS variant 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-03-17 02:53:58 +01:00
Rafael Ravedutti
887f41871c Add parameter reading for LAMMPS variant 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-03-17 02:44:34 +01:00
rafaelravedutti
459853dc25 Merge pull request #4 from RRZE-HPC/gromacs_sp 
Gromacs sp
 2022-03-15 20:31:42 +01:00
Rafael Ravedutti
d61576699d Add first compilable version of Gromacs with SP 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
 2022-03-15 02:40:56 +01:00
Jan Eitzinger
c2fcd50773 Initial version of lammps halfneighbor list 2022-03-10 17:06:45 +01:00