On 8/16/07, George Bosilca wrote:
> Well, finally someone discovered it :) I know about this problem for
> quite a while now, it pop up during our own valgrind test of the
> collective module in Open MPI. However, it never create any problems
> in the applications, at least not as far as I know. T
Hi, everybody.
I just added OpenMPI 1.2.3 to a Rocks cluster that already had OpenMPI
1.1 installed. I built an rpm for 1.2.3 from the tarball I had
downloaded from open-mpi.org, and I used the option --with-devel-
headers, which I needed in order to build a software package that
required all the
Thanks this is all interesting articles.
Brock Palen
Center for Advanced Computing
bro...@umich.edu
(734)936-1985
On Aug 17, 2007, at 3:13 PM, Jeff Squyres wrote:
On Aug 17, 2007, at 12:16 PM, Brock Palen wrote:
We have a user who uses the sepran1206 package. It works for him on
lam, mpich
On Aug 17, 2007, at 12:16 PM, Brock Palen wrote:
We have a user who uses the sepran1206 package. It works for him on
lam, mpich2 and OMPI up to problem sizes i see in the debugger (ddt)
that both rank 0 and rank 1 call PMPI_SEND()
Is PMPI_SEND the same as MPI_SEND?
For the most part, yes. In
If you care, there's a decent explanation on that web page of what
the issue is and why MPI has problems with it (mainly dealing with
non-blocking communication operations), and another good explanation
is provided in the 2nd half of chapter 10 of the MPI-2 spec (the
Fortran section of the
Dear George,
I think that the best way is to call MPI_Abort. However, this forces the
user to modify the code, which I already have suggested. But their
application is not calling exit directly, I merely wrote the simplest code
that demonstrates the problem. Their application is a Fortran p
This is not a problem generated by the MPI library. The problem come
from the Fortran compilers themselves and therefore apply to all MPI
libraries (including Open MPI).
george.
On Aug 17, 2007, at 12:34 PM, Brock Palen wrote:
As a SA I am not well learned in the ways of fortran, I am al
As a SA I am not well learned in the ways of fortran, I am also less
learned in the ways of the internals of OMPI.
We use the PGI compilers with OMPI and we have a user who is
wondering does the the following issue apply to openMPI?
http://nf.apac.edu.au/training/MPI/MPI_2.0/node236.html#Node
The MPI standard state that the correct way to abort/kill an MPI
application is using the MPI_Abort function. Except, if you're doing
some kind of fault tolerance stuff, there is no reason to end one of
your MPI processes via exit.
Thanks,
george.
On Aug 16, 2007, at 12:04 PM, Daniel
We have a user who uses the sepran1206 package. It works for him on
lam, mpich2 and OMPI up to problem sizes i see in the debugger (ddt)
that both rank 0 and rank 1 call PMPI_SEND()
Is PMPI_SEND the same as MPI_SEND?
also why would it work with lam and mpich2 ?
If we up the btl_sm_eager_lim
10 matches
Mail list logo
