Fix BSSN C gauge RHS parity

Shell C functions must export Fortran-compatible symbols with trailing
underscore so bssn_rhs_ss.f90 and getnp4.f90 can link when WithShell is
active and USE_CXX_SHELL_KERNELS=1 replaces Fortran diff_new_sh.o.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

AMSS_NCKU_source/makefile

@@ -58,14 +58,10 @@ POLINT6_FLAG = -DPOLINT6_USE_BARYCENTRIC=$(POLINT6_USE_BARY)
 TRANSFER_CACHE_FLAG = -DBSSN_USE_TRANSFER_CACHE=$(EFFECTIVE_USE_TRANSFER_CACHE)
 ESCALAR_KERNEL_FLAG = -DBSSN_USE_ESCALAR_C_KERNEL=$(EFFECTIVE_USE_CXX_ESCALAR_KERNEL)
 
+## AMD AOCC build flags optimized for EPYC Zen 4 (-march=znver4)
+##   make                        -> opt  (PGO-guided, maximum performance)
+##   make PGO_MODE=instrument    -> instrument (Phase 1: collect fresh profile data)
 
-## GCC build flags (optimized for x86-64-v4)
-## PGO disabled (used negative optimization on Intel; not tested on GCC)
-CXXAPPFLAGS = -O3 -march=x86-64-v4 -ffast-math -mfma -flto \
-              -Dfortran3 -Dnewc $(INTERP_LB_FLAGS) \
-              $(TRANSFER_CACHE_FLAG) $(ESCALAR_KERNEL_FLAG) $(EM_KERNEL_FLAG)
-f90appflags = -O3 -march=x86-64-v4 -ffast-math -mfma -flto \
-              -cpp $(POLINT6_FLAG)
 
 .SUFFIXES: .o .f90 .C .for .cu
 
@@ -73,11 +69,11 @@ f90appflags = -O3 -march=x86-64-v4 -ffast-math -mfma -flto \
 	$(f90) $(f90appflags) -c $< -o $@
 
 .C.o:
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 # ShellPatch.C uses OpenMP for setupintintstuff search loops
 ShellPatch.o: ShellPatch.C
-	${CXX} $(CXXAPPFLAGS) $(OMP_FLAG) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) $(OMP_FLAG) -c $< $(filein) -o $@
 
 .for.o:
 	$(f77) -c $< -o $@
@@ -87,59 +83,59 @@ ShellPatch.o: ShellPatch.C
 
 # C rewrite of BSSN RHS kernel and helpers
 bssn_rhs_c.o: bssn_rhs_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 fderivs_c.o: fderivs_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 fdderivs_c.o: fdderivs_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 kodiss_c.o: kodiss_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 lopsided_c.o: lopsided_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 lopsided_kodis_c.o: lopsided_kodis_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 # C rewrite of shell-patch derivative kernels
 fderivs_sh_c.o: fderivs_sh_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 fdderivs_sh_c.o: fdderivs_sh_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 fderivs_shc_c.o: fderivs_shc_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 fdderivs_shc_c.o: fdderivs_shc_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 kodiss_sh_c.o: kodiss_sh_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 
 bssn_em_rhs_c.o: bssn_em_rhs_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 z4c_rhs_c.o: z4c_rhs_c.C
-	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 #interp_lb_profile.o: interp_lb_profile.C interp_lb_profile.h
-#	${CXX} $(CXXAPPFLAGS) -c $< $(filein) -o $@
+#	$(CXX) $(CXXAPPFLAGS) -c $< $(filein) -o $@
 
 ## TwoPunctureABE uses fixed optimal flags with its own PGO profile, independent of CXXAPPFLAGS
 TP_PROFDATA = /home/$(shell whoami)/AMSS-NCKU/pgo_profile/TwoPunctureABE.profdata
 TP_OPTFLAGS = -O3 -xHost -fp-model fast=2 -fma -ipo \
               -fprofile-instr-use=$(TP_PROFDATA) \
-              -Dfortran3 -Dnewc $(filein_real)
+              -Dfortran3 -Dnewc -I$(AOCL_ROOT)/include
 
 TwoPunctures.o: TwoPunctures.C
-	${CXX} $(TP_OPTFLAGS) -qopenmp -c $< -o $@
+	$(CXX) $(TP_OPTFLAGS) -qopenmp -c $< -o $@
 
 TwoPunctureABE.o: TwoPunctureABE.C
-	${CXX} $(TP_OPTFLAGS) -qopenmp -c $< -o $@
+	$(CXX) $(TP_OPTFLAGS) -qopenmp -c $< -o $@
 
 # Input files
 

AMSS_NCKU_source/makefile.inc

@@ -1,26 +1,21 @@
-## GCC version with OpenMPI and OpenBLAS
+## AMD AOCC version with AOCL (Optimized for AMD EPYC Zen 4)
-OMPI_ROOT    = /usr/mpi/gcc/openmpi-4.1.9a1
 
-## Ensure mpicxx and final executables find OpenMPI libs at build- and runtime
+## AOCL root path for includes and libraries
-export LD_LIBRARY_PATH := $(OMPI_ROOT)/lib64:$(LD_LIBRARY_PATH)
+AOCL_ROOT ?= /home/aocc/aocl/5.2.0/aocc
 
-filein  = -I/usr/include/ -I$(OMPI_ROOT)/include
+## AOCC-built OpenMPI prefix
+OMPI_PREFIX ?= /home/aocc/openmpi-5.0.10
 
-## OpenBLAS (OpenMP variant) + gfortran runtime
+filein  = -I/usr/include/ -I$(AOCL_ROOT)/include
-## -Wl,-rpath ensures ABE / TwoPunctureABE find libmpi at runtime without LD_LIBRARY_PATH
+
-LDLIBS  = -Wl,-rpath,$(OMPI_ROOT)/lib64 -lopenblaso -lgfortran -lpthread -lm -ldl -lgomp
+## Using AOCL BLIS + libFLAME for BLAS/LAPACK
+## AOCC Fortran runtime: -lflang (includes FortranRuntime)
+## AOCC OpenMP runtime: -lomp (LLVM OpenMP)
+LDLIBS  = -L$(AOCL_ROOT)/lib -lblis -lflame -lamdlibm -lflang -lpgmath -lpthread -lm -ldl -lomp
 
 # OpenMP flag for selective compilation
 OMP_FLAG = -fopenmp
 
-## Memory allocator switch
-##   0 (default) : use system default allocator (ptmalloc)
-##   1           : use jemalloc (install jemalloc-devel first)
-USE_JEMALLOC ?= 0
-ifeq ($(USE_JEMALLOC),1)
-LDLIBS := -ljemalloc $(LDLIBS)
-endif
-
 ## Interp_Points load balance profiling mode
 ##   off        : (default) no load balance instrumentation
 ##   profile    : Pass 1 — instrument Interp_Points to collect timing profile
@@ -68,11 +63,11 @@ USE_TRANSFER_CACHE ?= auto
 ##   0           : use original Fortran rungekutta4_rout.o
 USE_CXX_RK4 ?= 1
 
-f90          = gfortran
+f90          = flang
-f77          = gfortran
+f77          = flang
-CXX          = g++
+CXX          = clang++
-CC           = gcc
+CC           = clang
-CLINKER      = mpicxx
+CLINKER      = $(OMPI_PREFIX)/bin/mpicxx
 
 Cu = nvcc
 CUDA_LIB_PATH = -L/usr/lib/cuda/lib64 -I/usr/include -I/usr/lib/cuda/include