Hello,
I have downloaded OpenMPI-1.8.1 and compiled it using Intel Compilers
(Intel Composer XE Suites 2013) and the command used is as follow:
[Code]
./configure --prefix=/opt/openmpi-1.8.1 CC=icc CXX=icpc F77=ifort FC=ifort
make all install
[/code]
Everything works just fine except I realized that the MPI library was built
for 32-bit integers (fort integer size=4)
Then I reconfigured it using this command:
[Code]
./configure --prefix=/opt/openmpi-1.8.1 CC=icc CXX=icpc F77=ifort FC=ifort
FFLAGS=-i8 FCFLAGS=-i8
[/code]
However, when I tried to compile it using the make all install command,
some errors appeared at the end of the compilation which is described as
followed:
[qoute]
... ...
CC pdist_graph_create_adjacent_f.lo
pdist_graph_create_adjacent_f.c(79): error: "c_sourceweights" has already
been declared in the current scope
OMPI_ARRAY_NAME_DECL(sourceweights);
^
pdist_graph_create_adjacent_f.c(81): error: "c_destweights" has already
been declared in the current scope
OMPI_ARRAY_NAME_DECL(destweights);
^
compilation aborted for pdist_graph_create_adjacent_f.c (code 2)
make[3]: *** [pdist_graph_create_adjacent_f.lo] Error 1
make[3]: Leaving directory
`/opt/openmpi-1.8.1/ompi/mpi/fortran/mpif-h/profile'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/opt/openmpi-1.8.1/ompi/mpi/fortran/mpif-h'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/opt/openmpi-1.8.1/ompi'
make: *** [all-recursive] Error 1
[/qoute]
I have no idea of what is happening. Would anyone please help me to solve
the problem if it is possible?
Thank you very much.
