Hi,
I have a simple program which performs the following code:
MPI_Init( &argc, &argv );
MPI_Errhandler_set(MPI_COMM_WORLD, MPI_ERRORS_RETURN);
MPI_Open_port(MPI_INFO_NULL, port_name);
printf("accepting connections on port: %s\n", port_name);
(... then the program MPI_Comm_accept)
I run the progr
Hi,
the port number which you see here is the correct format. Please note,
that Open MPI can only connect two applications within the same
"universe", and therefore we do not use hostnames or IP-Addresses in the
port name. For two applications to be in the same universe, you have to
start ort
On Apr 28, 2006, at 1:39 PM, Bernard Knaepen wrote:
Hello,
I am trying to install/run open-mpi on a Macbook Pro running MacOSX
10.4.6, *with* fortran support.
I am using Intel Fortran Compiler 9.1 (professional edition).
Compilation/installation went fine, except that the ifort compiler was
not
Hello All-
I am having a problem running ./configure for the OpenMPI 1.02
install. It is a problem with the Intel C++ compiler (icpc) -- I am
using Intel compilers:
Intel(R) C++ Compiler for 32-bit applications, Version 9.0Build
20051201Z Package ID: l_cc_c_9.0.030
Intel(R) C Compile
Hi,
I've been using open-mpi in Solaris for a while and now I need to use
open-mpi in AIX.
I compiled the latest stable open-mpi version and run one of the
simplest possible tests (see hello0.c in attached zip)
To my surprise, it didn't show anything to the screen when running with
mpirun:
mai
Hi,
I tried to compile a FORTRAN 77 test case and it failed when linking:
%>mpif77 -o fpi fpi.f
"/usr/local/include/mpif.h", line 64.38: 1515-009 (E) Null literal
string is not permitted. A single blank is assumed.
** main === End of Compilation 1 ===
1501-510 Compilation successful for file
On May 1, 2006, at 1:58 PM, Brignone, Sergio wrote:
I've been using open-mpi in Solaris for a while and now I need to use
open-mpi in AIX.
I compiled the latest stable open-mpi version and run one of the
simplest possible tests (see hello0.c in attached zip)
To my surprise, it didn't show anyt
I am in fact trying to use Open Mpi. In order to prepare my previous
email, I saved the different outputs but from the wrong
directory...sorry!
I attached here the correct files.
I also made sure today I am not mixing LAM/MPI and Open MPI.
Thanks,
Bernard.
On 5/1/06, Brian Barrett wrote:
On
Hi Brian,
I tried what you suggested (piping out stdout) but it did't work.
Sergio
-Original Message-
From: Brian Barrett [mailto:brbar...@open-mpi.org]
Sent: Monday, May 01, 2006 11:14 AM
To: Open MPI Users
Subject: Re: [OMPI users] stdout in AIX
On May 1, 2006, at 1:58 PM, Brignone,
I've noticed that I can't just fix this myself, very bad things
happened to the merged communicator, so this is not a trivial fix I
gather.
Michael
On Apr 30, 2006, at 12:16 PM, Michael Kluskens wrote:
MPI_Intercomm_Merge( intercomm, high, newintracomm, ier )
None of the books I have stat
checking if FORTRAN compiler supports integer(selected_int_kind
(2))... yes
checking size of FORTRAN integer(selected_int_kind(2))... unknown
configure: WARNING: *** Problem running configure test!
configure: WARNING: *** See config.log for details.
configure: error: *** Cannot continue.
Source
I get openmpi-1.0.2 to compile on a (small) G5 cluster. The C and C++
compilers work fine so far, but the mpif77 and mpif90 scripts send
the wrong flags to the f77 and f90 compilers. I've attached the
config.log , make.out, the error
messages, and the environment variables.
for f77 I get
Some more hints to the issue:
If I start process A and this process spawns processes B and C (using
MPI_Comm_spawn), stdout works fine only for process A.
Sergio
-Original Message-
From: Brignone, Sergio
Sent: Monday, May 01, 2006 1:17 PM
To: Open MPI Users
Subject: Re: [OMPI users] s
You're absolutely right -- this is another logical type that we missed.
This has now been fixed on both the trunk and the v1.1 branch; should be
in tomorrow's snapshots.
Thanks!
> -Original Message-
> From: users-boun...@open-mpi.org
> [mailto:users-boun...@open-mpi.org] On Behalf Of M
I just fixed the INTERCOMM_MERGE/logical issue on the trunk and the v1.1
branch -- can you give it a whirl there?
I ask because this issue is a bug that we fixed on the trunk (and
therefore v1.1) and didn't back-port it to v1.0. There's actually quite
a few of these F90 fixes on the trunk/v1.1 br
15 matches
Mail list logo