Compiling in AMD Opteron using G95
LIANG Yunfeng
liangy... at gmail.com
Fri Mar 14 23:21:28 UTC 2008
To be more precise, I will report the problem below. Even it is not related
to cp2k yet. It is just a matter of Lapack...Sorry for this, but all of you
may have suggestion then!!See below:
Using 64 bit g95 compiler, I can compile lapack-3.1.1 linked with
GotoBLAS-1.06, while I can't compile lapack-3.1.1 linked with
GotoBLAS-1.24-----this is a bug! Using 32 bit g95 compiler, I could not
compile lapack-3.1.1 linked with GotoBLAS-1.24 as well. So I should compile
lapack-3.1.1 linked with GotoBLAS-1.06 by the 32-bit G95 compiler "Linux
x86_64/EMT64 (32 bit D.I.)". But the "make" doesn't realize the command in
this case! At the bottom, I attached the "Makefile.rule" of GotoBLAS, which
in practice gives the following error message...
**********************************************************
make[2]: Leaving directory `/home/liang/library/GotoBLAS-1.06/level1/others'
make[2]: Entering directory `/home/liang/library/GotoBLAS-1.06/level1/rot'
c -O2 srot.f -o srot.o
make[2]: c: Command not found
make[2]: [srot.o] Error 127 (ignored)
c -O2 drot.f -o drot.o
make[2]: c: Command not found
make[2]: [drot.o] Error 127 (ignored)
c -O2 csrot.f -o csrot.o
make[2]: c: Command not found
make[2]: [csrot.o] Error 127 (ignored)
c -O2 zdrot.f -o zdrot.o
make[2]: c: Command not found
make[2]: [zdrot.o] Error 127 (ignored)
ar -ruv ../../libgoto_opteron-r1.06.a srot.o drot.o csrot.o zdrot.o
ar: srot.o: No such file or directory
***********************************************************
I addition, I found that similiar Makefile.rule of GotoBLAS works in
compiling GotoBLAS-1.24. It doesn't work just when you compile
lapack-3.1.1and cp2k. I don't know how to solve the problem. Could
you?
Thanks!
Sincerely, Yunfeng
*****************************
Makefile.rule**************************************
#
# Beginning of user configuration
#
# This library's version
REVISION = -r1.06
# Which do you prefer to use for C compiler? Default is gcc.
# I recommend you to use GCC because inline assembler is required.
# C_COMPILER = GNU
# C_COMPILER = INTEL
# Which do you prefer to use for Fortran compiler? Default is GNU G77.
# F_COMPILER = G77
# F_COMPILER = G95
F_COMPILER = /home/liang/g95-install/bin/x86_64-unknown-linux-gnu-g95
# F_COMPILER = GFORTRAN
# F_COMPILER = INTEL
# F_COMPILER = PGI
# F_COMPILER = PATHSCALE
# F_COMPILER = IBM
# F_COMPILER = COMPAQ
# F_COMPILER = SUN
# F_COMPILER = F2C
# If you want to build threaded version.
# You can specify number of threads by environment value
# "OMP_NUM_THREADS", otherwise, it's automatically detected.
# SMP = 1
# You may specify Maximum number of threads. It should be minimum.
# MAX_THREADS = 4
# If you need 64bit binary; some architecture can accept both 32bit and
# 64bit binary(EM64T, Opteron, SPARC and Power/PowerPC).
# BINARY64 = 1
# If you want to drive whole 64bit region by BLAS. Not all Fortran
# compiler supports this. It's safe to keep comment it out if you
# are not sure.
# INTERFACE64 = 1
# If you need Special memory management;
# Using HugeTLB file system(Linux / AIX / Solaris)
# CCOMMON_OPT += -DALLOC_HUGETLB
# Using static allocation instead of dynamic allocation
# You can't use it with ALLOC_HUGETLB together
# CCOMMON_OPT += -DALLOC_STATIC
# If you want to use CPU affinity
# CCOMMON_OPT += -DUSE_CPU_AFFINITY
# If you want to use memory affinity (for NUMA)
# You can't use it with ALLOC_STATIC together
# NUMA_AFFINITY = 1
# If you want to use pure thread server model.
# Default is only OMP_NUM_THREADS - 1 threads are spawned to reduce
# thread overhead. This is not implemented yet.
# CCOMMON_OPT += -DALL_THREADED
# Use busy wait instead of pthread_lock implemenation
# BLAS perforamnce should be better, but total performance would be
# worse due to extra usage of CPU
# CCOMMON_OPT += -DUSE_BUSYWAIT
# If you have special compiler to run script to determine architecture.
GETARCH_CC =
GETARCH_FLAGS =
#
# End of user configuration
#
....
....
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20080314/706138f3/attachment.htm>
More information about the CP2K-user
mailing list