On Thu, Aug 02, 2007 at 02:31:30AM +, Dirk Eddelbuettel wrote:
> Dear Open MPI developers,
Hi!
> We (as in the Debian maintainer for Open MPI) got this bug report from
> Uwe who sees mpi apps segfault on Debian systems with the FreeBSD
> kernel.
> Any input would be greatly appreciated!
Uw
Dirk Eddelbuettel schrieb:
Dear Open MPI developers,
We (as in the Debian maintainer for Open MPI) got this bug report from
Uwe who sees mpi apps segfault on Debian systems with the FreeBSD
kernel. Is there anybody here familiar with any BSD peculiarties that
may play a role?
Any input would
my send finger was too fast, I wanted to add that according to the MPI
standard, MPI_Comm_rank is defined as
int MPI_Comm_rank ( MPI_Comm comm, int *rank )
therefore I wouldn't recommend to use anything else than an int for rank...
regards, Bettina
Dirk Eddelbuettel schrieb:
Dear Open MPI de
Hi everybody,
News from the vampir trace integration at last. The current vampir trace
release is included in the TMP tree of the OpenMPI SVN. Now, autoconf scripts
need to be added.
Then the build procedure will look like this:
1) run global OpenMPI configure script
2) the global configure
Hi,
On Thu, Aug 02, 2007 at 02:13:22PM +0200, Bettina Krammer wrote:
> my send finger was too fast, I wanted to add that according to the MPI
> standard, MPI_Comm_rank is defined as
> int MPI_Comm_rank ( MPI_Comm comm, int *rank )
> therefore I wouldn't recommend to use anything else than an int
Hi,
On Thu, Aug 02, 2007 at 10:51:13AM +0200, Adrian Knoth wrote:
> Uwe, could you please recompile with --enable-debug and rerun the test?
> If possible, also provide a gdb backtrace, probably with details about
> the failing frame.
OK, I'll recompile with --enable-debug and see what happens.
Once you attach with gdb to the mpirun process, before getting the
stack information please type "shared". I want to be sure that gdb
will correctly load the symbols from all shared libraries (which is
not the default everywhere). The previously reported stack, doesn't
really seems consiste
On Aug 2, 2007, at 9:38 AM, Uwe Hermann wrote:
Hi,
On Thu, Aug 02, 2007 at 10:51:13AM +0200, Adrian Knoth wrote:
Uwe, could you please recompile with --enable-debug and rerun the
test?
If possible, also provide a gdb backtrace, probably with details
about
the failing frame.
OK, I'll reco
The modifications:
* to remove the #ifs
* put the sparse paramters in structs, and a union of them to the group
struct
are all there in the branch since yesterday. I tested the branch with the
ibm and intel test suites again, using all the different formats when
possible, and they passed.
I just