Hi Pradeep Just curious:
Did you change "mpi_int" into "MPI_INTEGER" also? Presumably the two types are the same size, but I wonder if somehow the two type names are interchangeable in OpenMPI (I would guess they're not), although declared in different header files, etc. Thank you, Gus Correa On 02/18/2013 08:32 PM, Pradeep Jha wrote:
That was careless of me. Thanks for pointing it out. Declaring "status", "ierr" and putting "implicit none" solved the problem. Thanks again. 2013/2/19 Jeff Squyres (jsquyres) <jsquy...@cisco.com <mailto:jsquy...@cisco.com>> +1. The problem is that you didn't declare status or ierr. Since you didn't declare status, you're buffer overflowing, and random Bad Things happen from there. You should *always* "implicit none" to catch these kinds of errors. On Feb 18, 2013, at 2:02 PM, Gus Correa <g...@ldeo.columbia.edu <mailto:g...@ldeo.columbia.edu>> wrote: > Hi Pradeep > > For what it is worth, in the MPI Fortran bindings/calls the > datatype to use is "MPI_INTEGER", not "mpi_int" (which you used; > MPI_INT is in the MPI C bindings): > > http://linux.die.net/man/3/mpi_integer > > Also, just to prevent variables to inadvertently come with > the wrong type, you could add: > > implicit none > > to the top of your code. > You already have a non-declared "ierr" in "call mpi_send". > (You declared "ierror" as an integer, but not "ierr".) > Although this one may not cause any harm; > names starting with "i" are integers by default, in old Fortran. > > I hope this helps, > Gus Correa > > > On 02/18/2013 01:26 PM, jody wrote: >> Hi Pradeep >> >> I am not sure if this is the reason, but usually it is a bad idea to >> force an order of receives (such as you do in your receive loop - >> first from sender 1 then from sender 2 then from sender 3) >> Unless you implement it so, there is no guarantee the sends are >> performed in this order. B >> >> It is better if you accept messages from all senders (MPI_ANY_SOURCE) >> instead of particular ranks and then check where the >> message came from by examining the status fields >> (http://www.mpi-forum.org/docs/mpi22-report/node47.htm) >> >> Hope this helps >> Jody >> >> >> On Mon, Feb 18, 2013 at 5:06 PM, Pradeep Jha >> <prad...@ccs.engg.nagoya-u.ac.jp <mailto:prad...@ccs.engg.nagoya-u.ac.jp>> wrote: >>> I have attached a sample of the MPI program I am trying to write. When I run >>> this program using "mpirun -np 4 a.out", my output is: >>> >>> Sender: 1 >>> Data received from 1 >>> Sender: 2 >>> Data received from 1 >>> Sender: 2 >>> >>> And the run hangs there. I dont understand why does the "sender" variable >>> change its value after MPI_recv? Any ideas? >>> >>> Thank you, >>> >>> Pradeep >>> >>> >>> program mpi_test >>> >>> include 'mpif.h' >>> >>> !----------------( Initialize variables )-------------------- >>> integer, dimension(3) :: recv, send >>> >>> integer :: sender, np, rank, ierror >>> >>> call mpi_init( ierror ) >>> call mpi_comm_rank( mpi_comm_world, rank, ierror ) >>> call mpi_comm_size( mpi_comm_world, np, ierror ) >>> >>> !----------------( Main program )-------------------- >>> >>> ! receive the data from the other processors >>> if (rank.eq.0) then >>> do sender = 1, np-1 >>> print *, "Sender: ", sender >>> call mpi_recv(recv, 3, mpi_int, sender, 1, >>> & mpi_comm_world, status, ierror) >>> print *, "Data received from ",sender >>> end do >>> end if >>> >>> ! send the data to the main processor >>> if (rank.ne.0) then >>> send(1) = 3 >>> send(2) = 4 >>> send(3) = 4 >>> call mpi_send(send, 3, mpi_int, 0, 1, mpi_comm_world, ierr) >>> end if >>> >>> >>> !----------------( clean up )-------------------- >>> call mpi_finalize(ierror) >>> >>> return >>> end program mpi_test` >>> >>> >>> _______________________________________________ >>> users mailing list >>> us...@open-mpi.org <mailto:us...@open-mpi.org> >>> http://www.open-mpi.org/mailman/listinfo.cgi/users >> _______________________________________________ >> users mailing list >> us...@open-mpi.org <mailto:us...@open-mpi.org> >> http://www.open-mpi.org/mailman/listinfo.cgi/users > > _______________________________________________ > users mailing list > us...@open-mpi.org <mailto:us...@open-mpi.org> > http://www.open-mpi.org/mailman/listinfo.cgi/users -- Jeff Squyres jsquy...@cisco.com <mailto:jsquy...@cisco.com> For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ _______________________________________________ users mailing list us...@open-mpi.org <mailto:us...@open-mpi.org> http://www.open-mpi.org/mailman/listinfo.cgi/users _______________________________________________ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users
