JanLJL
|
95d63334fa
|
added asm files and analysis output
|
2023-02-13 14:15:08 +01:00 |
|
JanLJL
|
d0277765c3
|
changed double constants to MD_FLOAT constants
|
2023-02-13 14:13:53 +01:00 |
|
Rafael Ravedutti
|
5814a86125
|
Small fixes
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-02-12 01:44:48 +01:00 |
|
JanLJL
|
98583cdade
|
Merge branch 'master' of github.com:RRZE-HPC/MD-Bench
|
2023-02-09 17:33:42 +01:00 |
|
JanLJL
|
cb5598bc91
|
added static analysis and likwid files
|
2023-02-09 17:33:22 +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 |
|
Rafael Ravedutti
|
32e004944f
|
Fix flags parsing for script
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-23 22:54:44 +01:00 |
|
Rafael Ravedutti
|
6126d74aa9
|
Change latency and CFG script to use binary as input
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-23 22:50:37 +01:00 |
|
Rafael Ravedutti
|
016f07dcaa
|
Merge branch 'master' of github.com:RRZE-HPC/MD-Bench
|
2023-01-23 22:13:42 +01:00 |
|
Rafael Ravedutti
|
90f30d26a3
|
Add lammps-avx512 kernels without correction instructions
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-23 22:13:33 +01:00 |
|
rafaelravedutti
|
01cc05a5d6
|
Merge pull request #5 from scarboot/master
Move likwid marker calls into OpenMP parallel region
|
2023-01-22 17:20:02 +01:00 |
|
Yannick Paschke
|
c61cf9a0ac
|
Move likwid marker calls into OpenMP parallel region
|
2023-01-22 15:33:05 +01:00 |
|
Rafael Ravedutti
|
d545ca65d4
|
Add -xHost option for AVX2
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-18 16:39:19 +01:00 |
|
Rafael Ravedutti
|
5833f00894
|
Change ICX flags based on ISA
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-16 23:13:40 +01:00 |
|
Rafael Ravedutti
|
8aad7e87a0
|
Use ISA for GCC flags and change binary and build directory names
Signed-off-by: Rafael Ravedutti <rafael.r.ravedutti@fau.de>
|
2023-01-16 23:05:21 +01:00 |
|
Rafael Ravedutti
|
ffad9d40f3
|
Use ICC compiler options based on ISA
Signed-off-by: Rafael Ravedutti <rafael.r.ravedutti@fau.de>
|
2023-01-16 22:45:24 +01:00 |
|
Rafael Ravedutti
|
99da76d59c
|
Add flags with -march=core-avx2 for Milan
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-11 15:30:26 +01:00 |
|
Rafael Ravedutti
|
cfe888c132
|
Add analysis files from gromacs-avx512-dp with ICX compiler
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-03 16:14:28 +01:00 |
|
Rafael Ravedutti
|
c7b136f629
|
Fix build directories in .gitignore
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-03 16:13:26 +01:00 |
|
Rafael Ravedutti
|
07f2f74561
|
Adjust force_iters stats for 4xN kernel
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-02 23:57:51 +01:00 |
|
Rafael Ravedutti
|
fd368609e8
|
Remove binaries on cleanall
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-02 19:15:42 +01:00 |
|
Rafael Ravedutti
|
db5f8cf1c6
|
Update .gitignore
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-02 19:12:25 +01:00 |
|
Rafael Ravedutti
|
f467d10ed3
|
Add cleanall command for all TAG and OPT_SCHEME variants
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-02 19:11:05 +01:00 |
|
Rafael Ravedutti
|
fe86c948a8
|
Adjust time and likwid measurements on 4xN kernels
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-02 14:19:59 +01:00 |
|
Rafael Ravedutti
|
ae1cfa2800
|
Include static_analysis directory
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2023-01-02 14:14:16 +01:00 |
|
Rafael Ravedutti
|
e5c233e072
|
Update script
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-21 18:04:18 +01:00 |
|
Rafael Ravedutti
|
8d5e10f635
|
Fix compilation for gromacs-avx512-sp
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-21 16:19:00 +01:00 |
|
Rafael Ravedutti
|
56ff0d19af
|
Run script with different prefetcher settings and provide better output
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-20 18:51:54 +01:00 |
|
Rafael Ravedutti
|
b20e8c6986
|
Adjust script for GROMACS scheme
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-14 17:54:18 +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
|
292138b270
|
Write debug_printf to avoid warnings and fix latency/cfd script
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-14 16:17:28 +01:00 |
|
Rafael Ravedutti
|
0e952964f7
|
Update script
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-13 15:47:17 +01:00 |
|
Rafael Ravedutti
|
cd37746d07
|
Add script to automate latency and CFD evaluation
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-13 15:35:41 +01:00 |
|
Rafael Ravedutti
|
d76d044a00
|
Update build options for each compiler and include ICX
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-13 01:06:59 +01:00 |
|
Rafael Ravedutti
|
a7cb888517
|
Fix compilation for ICX compiler
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-12-13 00:59:31 +01:00 |
|
Rafael Ravedutti
|
416f042fc0
|
Fix readline() when fgets returns NULL even on success
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-30 17:43:35 +01:00 |
|
Rafael Ravedutti
|
fa4e38c6c4
|
Add IACA and stubbed measurements for GROMACS 4x8 FN kernel
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-18 01:00:20 +01:00 |
|
Rafael Ravedutti
|
04ea1b027e
|
Print kernel and precision info in gromacs-stub
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-16 16:15:15 +01:00 |
|
Rafael Ravedutti
|
56d9613028
|
Implement stubbed version for GROMACS
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-15 16:01:13 +01:00 |
|
Rafael Ravedutti
|
bc06220aeb
|
Remove AVX512 reciprocal usage in AVX2 file
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-15 01:40:37 +01:00 |
|
Rafael Ravedutti
|
efa462d0af
|
Add AVX_FMA ISA
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-15 01:24:30 +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
|
f293cec960
|
Call CPU version of updatePbc within setupPbc
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-14 19:19:57 +01:00 |
|
Rafael Ravedutti
|
6eedf1776e
|
Small fixes into GROMACS GPU code
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-14 18:21:14 +01:00 |
|
Rafael Ravedutti
|
93188d1383
|
Adjust NVCC flags to avoid issues with atomicAdd with doubles
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-11-14 18:01:46 +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 |
|
Rafael Ravedutti
|
c4304e3619
|
Update figure widths again
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
|
2022-09-29 18:41:40 +02:00 |
|