Edgar, here is the F77 version of the source code.
Thanks
Sergio
-Original Message-
From: Edgar Gabriel [mailto:gabr...@cs.uh.edu]
Sent: Tuesday, March 07, 2006 12:09 PM
To: Open MPI Users
Subject: Re: [OMPI users] Spawn and Disconnect
I know that there was a bug in the F90 interface o
On Mar 7, 2006, at 3:23 PM, Michael Kluskens wrote:
Per the mpi_comm_spawn issues with the 1.0.x releases I started using
1.1r9212, with my sample code I'm getting a messages of
[-:13327] mca: base: component_find: unable to open: dlopen(/usr/
local/lib/openmpi/mca_pml_teg.so, 9): Symbol not fo
These errors indicate a mismatching between the components and the
mpi library. Please remove the installation directory and do a make
install again in the open mpi directory. Somewhere around the
revision 9100 we remove some of the components (teg included). This
error say that there is st
On Mar 7, 2006, at 8:45 AM, Michael Kluskens wrote:
I will begin using the 1.1 snapshot as soon as I see r9198 or
higher. The dynamic process creation is critical to my project and I
need the full range of features that have been discussed on the list
recently regarding the MPI_SPAWN.
I would
Per the mpi_comm_spawn issues with the 1.0.x releases I started using
1.1r9212, with my sample code I'm getting a messages of
[-:13327] mca: base: component_find: unable to open: dlopen(/usr/
local/lib/openmpi/mca_pml_teg.so, 9): Symbol not found:
_mca_ptl_base_recv_request_t_class
Refere
Hello again,
The problem is that MPI_SEND blocks forever (the message is still not
delivered after many hours).
Cezary Sliwa
--- Begin Message ---
My program runs fine with openmpi-1.0.1 when run from the command line
(5 processes with empty host file), but when I schedule it with qsub to
I know that there was a bug in the F90 interface of spawn-multiple,
however (which is fixed by now as far as I can tell). Could you send me
the f77 example which you have? The concatination problem looks strange,
I would like to have a look at it...
Thanks
Edgar
Brignone, Sergio wrote:
Than
Thanks Edgar, Ralph and Jean.
It seems to me that the problem I am having is related to the operating
system or MPI configuration or compiler or all of them (I am using
Solaris).
For example, the F90 as well as the C++ interfaces could not be compiled
(I had to configure MPI without them).
I co
I will begin using the 1.1 snapshot as soon as I see r9198 or
higher. The dynamic process creation is critical to my project and I
need the full range of features that have been discussed on the list
recently regarding the MPI_SPAWN.
I would think that errors like this should be backported
