Dear MPI Users and Maintainers,
I am using openMPI in version 1.10.4 with enabled multithread support and
java bindings. I use MPI in java, having one process per machine and
multiple threads per process.
I was trying to build a broadcast listener thread which calls MPI_iBcast,
followed by MPI_WA
Thank you very much. Could you please answer another somewhat related question?
I'd like to know if ORTE could be linked statically like a library in order to
have a completely stand-alone executable file. As you may have noticed I don't
have a good knowledge of how Open MPI works.
Thanks in adv
MPI 3.1 5.12 is pretty clear on the matter:
"It is erroneous to call MPI_REQUEST_FREE or MPI_CANCEL for a request
associated with a nonblocking collective operation."
-Nathan
> On Jun 9, 2017, at 5:33 AM, Markus wrote:
>
> Dear MPI Users and Maintainers,
>
> I am using openMPI in version 1.
Sure - just configure OMPI with “--enable-static --disable-shared”
> On Jun 9, 2017, at 5:50 AM, Arham Amouie via users
> wrote:
>
> Thank you very much. Could you please answer another somewhat related
> question? I'd like to know if ORTE could be linked statically like a library
> in order
It has been awhile since I tested it, but I believe the --enable-recovery
option might do what you want.
> On Jun 8, 2017, at 6:17 AM, Tim Burgess wrote:
>
> Hi!
>
> So I know from searching the archive that this is a repeated topic of
> discussion here, and apologies for that, but since it's
Hi. I had tried this. It still looks for ORTE file(s) on the hard disks of
compute nodes.
Now I know that I can install Open MPI in a shared directory. But is it
possible to make executable files that don't look for any Open MPI's files on
disk?
Arham
From: "r...@open-mpi.org"
To: Arh
Well, of course it still needs to execute the orteds on those nodes - but that
wasn’t what you asked. One way or another, the orteds must be available on the
compute nodes.
> On Jun 9, 2017, at 8:53 AM, Arham Amouie wrote:
>
> Hi. I had tried this. It still looks for ORTE file(s) on the hard d
Tim,
FT-MPI is gone, but the ideas it put forward have been refined and the
software algorithms behind them improved in a newer (and supported) project
ULFM. It features a smaller API, with a much more flexible approach. You
can find more information about it at http://fault-tolerance.org/. The
co
I was able to successfully install these versions. Thanks Ralph.
The error was path to 'libevent' getting overridden. A clean installation
of libevent, followed by PMIx and then OMPI went successfully.
On 7 June 2017 at 20:24, r...@open-mpi.org wrote:
> I guess I should also have clarified - I
