64-BitBrainstorm_2026/AMSS-NCKU
8
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Optimize numerical algorithms with Intel oneMKL

Browse Source 
- FFT.f90: Replace hand-written Cooley-Tukey FFT with oneMKL DFTI
   - ilucg.f90: Replace manual dot product loop with BLAS DDOT
   - gaussj.C: Replace Gauss-Jordan elimination with LAPACK dgesv/dgetri
   - makefile.inc: Add MKL include paths and library linking

   All optimizations maintain mathematical equivalence and numerical precision.
This commit is contained in:
CGH0S7
2026-01-16 10:58:11 +08:00
parent 7a76cbaafd
commit cb252f5ea2
4 changed files with 100 additions and 156 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
AMSS_NCKU_source/makefile.inc
View File

@@ -3,14 +3,14 @@
## filein = -I/usr/include/ -I/usr/include/openmpi-x86_64/ -I/usr/lib/x86_64-linux-gnu/openmpi/include/ -I/usr/lib/x86_64-linux-gnu/openmpi/lib/ -I/usr/lib/gcc/x86_64-linux-gnu/11/ -I/usr/include/c++/11/
## LDLIBS = -L/usr/lib/x86_64-linux-gnu -L/usr/lib64 -L/usr/lib/gcc/x86_64-linux-gnu/11 -lgfortran -lmpi -lgfortran
## Intel oneAPI version
filein = -I/usr/include/
## Intel oneAPI version with oneMKL
filein = -I/usr/include/ -I${MKLROOT}/include
LDLIBS = -L/usr/lib/x86_64-linux-gnu -L/usr/lib64 -lifcore -limf -lmpi
LDLIBS = -L/usr/lib/x86_64-linux-gnu -L/usr/lib64 -lifcore -limf -lmpi \
-L${MKLROOT}/lib -lmkl_intel_lp64 -lmkl_sequential -lmkl_core -lpthread -lm -ldl
CXXAPPFLAGS = -O3 -Dfortran3 -Dnewc
#f90appflags = -O3 -fpp
f90appflags = -O3 -fpp
CXXAPPFLAGS = -O3 -Dfortran3 -Dnewc -I${MKLROOT}/include
f90appflags = -O3 -fpp -I${MKLROOT}/include
f90 = ifx
f77 = ifx
CXX = icpx