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

Add OpenMP parallelization to Fortran compute kernels

Browse Source 
Add !$omp parallel do collapse(2) directives to all triple-loop
stencil kernels (fderivs, fdderivs, fdx/fdy/fdz, kodis, lopsided,
enforce_ag/enforce_ga) across all ghost_width variants. Add !$omp
parallel workshare to RK4/ICN/Euler whole-array update routines.

Build system: add -qopenmp to compile and link flags, switch MKL
from sequential to threaded (-lmkl_intel_thread -liomp5).

Runtime: set OMP_NUM_THREADS=96, OMP_STACKSIZE=16M, OMP_PROC_BIND=close,
OMP_PLACES=cores for 96-core server target.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
CGH0S7
2026-02-10 23:40:17 +08:00
parent caf192b2e4
commit 714c6e90c6
10 changed files with 171 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
AMSS_NCKU_source/rungekutta4_rout.f90
View File

@@ -109,23 +109,33 @@
if( RK4 == 0 ) then
!$omp parallel workshare
f1 = f0 + HLF * dT * f_rhs
!$omp end parallel workshare
elseif(RK4 == 1 ) then
!$omp parallel workshare
f_rhs = f_rhs + TWO * f1
!$omp end parallel workshare
!$omp parallel workshare
f1 = f0 + HLF * dT * f1
!$omp end parallel workshare
elseif(RK4 == 2 ) then
!$omp parallel workshare
f_rhs = f_rhs + TWO * f1
!$omp end parallel workshare
!$omp parallel workshare
f1 = f0 + dT * f1
!$omp end parallel workshare
elseif( RK4 == 3 ) then
!$omp parallel workshare
f1 = f0 +F1o6 * dT *(f1 + f_rhs)
!$omp end parallel workshare
else
@@ -134,7 +144,7 @@
endif
return
return
end subroutine rungekutta4_rout
!-----------------------------------------------------------------------------
@@ -215,15 +225,19 @@
if( RK4 == 0 ) then
!$omp parallel workshare
f1 = f0 + dT * f_rhs
!$omp end parallel workshare
else
!$omp parallel workshare
f1 = f0 + HLF * dT * (f1+f_rhs)
!$omp end parallel workshare
endif
return
return
end subroutine icn_rout
!~~~~~~~~~~~~~~~~~~
@@ -239,8 +253,10 @@
real*8, dimension(ex(1),ex(2),ex(3)),intent(in) ::f_rhs
real*8, dimension(ex(1),ex(2),ex(3)),intent(out) ::f1
!$omp parallel workshare
f1 = f0 + dT * f_rhs
!$omp end parallel workshare
return
return
end subroutine euler_rout