VMD-L Mailing List

From: surya narayanan chandrasekaran (sonofsunsurya_at_gmail.com)
Date: Tue Jun 11 2019 - 05:12:30 CDT

Hi,
     We are trying to use the Open quantum dynamics calculations with the
hierarchy equations of motion on parallel computers
<https://www.ks.uiuc.edu/Publications/Papers/abstract.cgi?tbcode=STRU2012A>
for our work. But we have trouble in installing the ATLAS tool by reducing
the CPU throttling. We use ubuntu bionic, in this ATLAS comes as default
lib. can anyone tell us the best possible way to compile the PHI program.

The make.config file is also confusing.
PHIVERSION=1.0

CC_LIN=g++
CC_SOL=gcc
CC_MAC=g++-4.2
CC_ICC=icc

#uncomment to enable change from double to single precision
#PRECISION=-DSINGLEPRECISION

#uncomment to enable thread affinity
#THREADAFFINITY=-DTHREADAFFINITY

CBLASINC_NOBLAS=-DNOBLAS
CBLASINC_LIN=-I/home/johanstr/apps/ATLAS3.8.3/include/
CBLASINC_MAC=-I/Developer/SDKs/MacOSX10.6.sdk/System/Library/Frameworks/vecLib.framework/Headers

CBLASLIB_NOBLAS=-DNOBLAS
CBLASLIB_LIN=-L/home/johanstr/apps/ATLAS3.8.3/bamako_build/lib/ -latlas
CBLASLIB_MAC=-latlas
CBLASLIB_SOL= /home/johanstr/apps/ATLAS3.8.3/CancunGCC_build/lib/libcblas.a
/home/johanstr/apps/ATLAS3.8.3/CancunGCC_build/lib/libatlas.a

CBLAS_STATIC=/home/johanstr/apps/ATLAS3.8.3/bamako_build/lib/libcblas.a
/home/johanstr/apps/ATLAS3.8.3/bamako_build/lib/libatlas.a

CFLAGS_NOBLAS=-Wall $(CBLASINC_NOBLAS)
CFLAGS_LIN=-Wall $(CBLASINC_LIN)
CFLAGS_MAC= -Wall $(CBLASINC_MAC) -DMAC
CFLAGS_SOL= -I/home/johanstr/local_sol/include $(CBLASINC) -DSOLARIS
CFLAGS_ICC= -Wall $(CBLASINC_LIN)

LDFLAGS_NOBLAS= $(CBLASLIB_NOBLAS) -lstdc++ -pthread -lpthread
LDFLAGS_LIN= $(CBLASLIB_LIN) -lstdc++ -pthread -lpthread
LDFLAGS_SOL= -L/usr/sfw/lib/64 -m64 -pthreads -lm -lc -lstdc++ -z muldefs
$(CBLASLIB_SOL)
LDFLAGS_MAC= -lstdc++ -lpthread $(CBLASLIB_MAC)
LDFLAGS_LIN_STATIC=-lstdc++ -pthread -lpthread

OPTIMISE_LIN= -O3 -pipe -fexpensive-optimizations -ffast-math
-funroll-all-loops -fprefetch-loop-arrays
OPTIMISE_MAC=-march=core2 -O3 -pipe -fexpensive-optimizations -ffast-math
-funroll-all-loops -fprefetch-loop-arrays
OPTIMISE_SOL= -m64 -O3 -pipe -fexpensive-optimizations -ffast-math
-mfpmath=sse -funroll-all-loops -fprefetch-loop-arrays
OPTIMISE_ICC= -O3 -pipe -fexpensive-optimizations -ffast-math
-funroll-all-loops -fprefetch-loop-arrays

CC=$(CC_LIN)
CFLAGS=$(CFLAGS_LIN)
LDFLAGS=$(LDFLAGS_LIN)
#CFLAGS=$(CFLAGS_NOBLAS)
#LDFLAGS=$(LDFLAGS_NOBLAS)
OPTIMISE=$(OPTIMISE_LIN) $(PRECISION) $(THREADAFFINITY)

#OPTIMISE=-g
LD=gcc 

-- 
Warm regards,
Dr. Suryanarayanan C