public inbox for cygwin@cygwin.com
 help / color / mirror / Atom feed
* Problem with OpenMPI
@ 2020-10-01 14:47 Rudnei Dias da Cunha
  2020-10-01 19:29 ` Marco Atzeri
  0 siblings, 1 reply; 3+ messages in thread
From: Rudnei Dias da Cunha @ 2020-10-01 14:47 UTC (permalink / raw)
  To: cygwin

Hi,
I have installed all the relevant OpenMPI packages to develop programs in
Fortran but found a problem when compiling programs in Fortran 90 with the
clause

use mpi

as mpifort then complains:

$ mpifort hello_usempi.f90
hello_usempi.f90:13:9:

   13 |     use mpi
      |         1
Fatal Error: Cannot open module file ‘mpi.mod’ for reading at (1): No such
file or directory

The workaround was to use

mpifort -J/usr/lib

as /usr/lib is the directory where mpi.mod resides after the installation
of the OpenMPI packages.

Has anyone found this before? Shouldn't the flag -J/usr/lib be issued as
part of the string issued by mpifort?

Regards
Rudnei

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Problem with OpenMPI
  2020-10-01 14:47 Problem with OpenMPI Rudnei Dias da Cunha
@ 2020-10-01 19:29 ` Marco Atzeri
       [not found]   ` <CAG0O4Lv+Hz8BnkWr+Wuo8bLAWihDwMJpvigNFtDsErthKdN3NA@mail.gmail.com>
  0 siblings, 1 reply; 3+ messages in thread
From: Marco Atzeri @ 2020-10-01 19:29 UTC (permalink / raw)
  To: cygwin

On 01.10.2020 16:47, Rudnei Dias da Cunha via Cygwin wrote:
> Hi,
> I have installed all the relevant OpenMPI packages to develop programs in
> Fortran but found a problem when compiling programs in Fortran 90 with the
> clause
> 
> use mpi
> 
> as mpifort then complains:
> 
> $ mpifort hello_usempi.f90
> hello_usempi.f90:13:9:
> 
>     13 |     use mpi
>        |         1
> Fatal Error: Cannot open module file ‘mpi.mod’ for reading at (1): No such
> file or directory
> 
> The workaround was to use
> 
> mpifort -J/usr/lib
> 
> as /usr/lib is the directory where mpi.mod resides after the installation
> of the OpenMPI packages.
> 
> Has anyone found this before? Shouldn't the flag -J/usr/lib be issued as
> part of the string issued by mpifort?
> 
> Regards
> Rudnei

there are not so many user of mpifort
but I do not see the problem building the examples:


$ make
mpicc -g  hello_c.c  -o hello_c
mpicc -g  ring_c.c  -o ring_c
mpicc -g  connectivity_c.c  -o connectivity_c
make[1]: Entering directory '/pub/devel/openmpi/examples64'
make[2]: Entering directory '/pub/devel/openmpi/examples64'
mpifort -g  hello_mpifh.f  -o hello_mpifh
mpifort -g  ring_mpifh.f  -o ring_mpifh
make[2]: Leaving directory '/pub/devel/openmpi/examples64'
make[2]: Entering directory '/pub/devel/openmpi/examples64'
mpifort -g  hello_usempi.f90  -o hello_usempi
mpifort -g  ring_usempi.f90  -o ring_usempi
make[2]: Leaving directory '/pub/devel/openmpi/examples64'
make[2]: Entering directory '/pub/devel/openmpi/examples64'
mpifort -g  hello_usempif08.f90  -o hello_usempif08
mpifort -g  ring_usempif08.f90  -o ring_usempif08
make[2]: Leaving directory '/pub/devel/openmpi/examples64'
make[1]: Leaving directory '/pub/devel/openmpi/examples64'


$ cat hello_usempi.f90
!
! Copyright (c) 2004-2006 The Trustees of Indiana University and Indiana
!                         University Research and Technology
!                         Corporation.  All rights reserved.
! Copyright (c) 2004-2005 The Regents of the University of California.
!                         All rights reserved.
! Copyright (c) 2006-2015 Cisco Systems, Inc.  All rights reserved.
! $COPYRIGHT$
!
! Sample MPI "hello world" application using the Fortran mpi module
! bindings.
!
program main
     use mpi
     implicit none
     integer :: ierr, rank, size, len
     character(len=MPI_MAX_LIBRARY_VERSION_STRING) :: version

     call MPI_INIT(ierr)
     call MPI_COMM_RANK(MPI_COMM_WORLD, rank, ierr)
     call MPI_COMM_SIZE(MPI_COMM_WORLD, size, ierr)
     call MPI_GET_LIBRARY_VERSION(version, len, ierr)

     write(*, '("Hello, world, I am ", i2, " of ", i2, ": ", a)') &
           rank, size, version

     call MPI_FINALIZE(ierr)
end




^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Problem with OpenMPI
       [not found]   ` <CAG0O4Lv+Hz8BnkWr+Wuo8bLAWihDwMJpvigNFtDsErthKdN3NA@mail.gmail.com>
@ 2020-10-03 15:12     ` Marco Atzeri
  0 siblings, 0 replies; 3+ messages in thread
From: Marco Atzeri @ 2020-10-03 15:12 UTC (permalink / raw)
  To: cygwin

back on the mailing list

On 03.10.2020 16:55, Massimiliano Alvioli wrote:
>>> Has anyone found this before? Shouldn't the flag -J/usr/lib be issued as
>>> part of the string issued by mpifort?
> 
> I found this a few times - not using mpifort but mpif90 - and I
> believe it may be due to
> a mismatch between the version of gcc and the compiler used to build mpi
> 
> 
> M
> 

interesting, but I do not see the problem and

$ gfortran  --version
GNU Fortran (GCC) 10.2.0

but when I built the openmpi-3.1.6 was last may with version 9


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2020-10-03 15:12 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-10-01 14:47 Problem with OpenMPI Rudnei Dias da Cunha
2020-10-01 19:29 ` Marco Atzeri
     [not found]   ` <CAG0O4Lv+Hz8BnkWr+Wuo8bLAWihDwMJpvigNFtDsErthKdN3NA@mail.gmail.com>
2020-10-03 15:12     ` Marco Atzeri

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).