That would be right, this is my NAMD_2.6_Source/arch/Linux-amd64-
MPI.arch:
NAMD_ARCH = Linux-amd64
CHARMARCH = mpi-linux-amd64
CXX = mpiCC
#CXXOPTS = -O3 -m64 -fexpensive-optimizations -ffast-math
CXXOPTS = -fastsse -O3 -Minfo -fPIC
#CC = gcc
CC = mpicc
#COPTS = -O3 -m64 -fexpensive-optimizations -ffast-math
COPTS = -fastsse -O3 -Minfo -fPIC
You will also want to edit:
NAMD_2.6_Source/arch/Linux-amd64.fftw
If you want it
Brock Palen
Center for Advanced Computing
bro...@umich.edu
(734)936-1985
On Aug 7, 2007, at 10:43 AM, Rainer Keller wrote:
Hello dear Brock and Graham,
we have had NAMD on our clusters with Open MPI. look for any file
conv-mach.sh; for the configuration mpi-linux, or in Your case
mpi-linux-amd64 this contains the superfluous -lmpich.
With best regards,
Rainer
On Tuesday 07 August 2007 04:11, Graham Jenkins wrote:
Brock Palen wrote:
I have done work before to make namd and charm++ work with openMPI I
dont remember what but it is doable. Something like removing -
lmpich
was enough i think, maybe a hack to use mpiCC and -fPIC (pgi
compilers).
I could look more if you want.
--
I'd really appreciate that Brock, thanks! Where would one remove
"-lmpich" from? I've had some difficulty finding it.
It actually builds OK using:
./build charm++ mpi-linux-amd64 ifort \
--basedir /opt/sw/openmpi-1.2.3-i
But if barfs when you try to do: "try out a sample program like
tests/charm++/simplearrayhello"
You can actually make the test compile by doing:
cd /opt/sw/openmpi-1.2.3-i/lib ; ln -s libmpi.so.0.0.0 libmpich.so
.. but I'm not sure that it's legit! :)
--
----------------------------------------------------------------
Dipl.-Inf. Rainer Keller http://www.hlrs.de/people/keller
High Performance Computing Tel: ++49 (0)711-685 6 5858
Center Stuttgart (HLRS) Fax: ++49 (0)711-685 6 5832
POSTAL:Nobelstrasse 19 email: kel...@hlrs.de
ACTUAL:Allmandring 30, R.O.030 AIM:rusraink
70550 Stuttgart
