Thanks Gilles!
That works with this MPI_Info given to MPI_Publish_name.
Cheers,
Bernard
Le 08/01/2015 03:47, Gilles Gouaillardet a écrit :
Well, per the source code, this is not a bug but a feature :
from publish function from ompi/mca/pubsub/orte/pubsub_orte.c
Thanks!
> On Jan 7, 2015, at 6:56 PM, Gilles Gouaillardet
> wrote:
>
> Ralph,
>
> honestly, i do not know if the standard says anything
> (all i can do for now is blame the coin ;-) )
>
> from now, i ll make a PR to update the man page (only master refers
>
Ralph,
honestly, i do not know if the standard says anything
(all i can do for now is blame the coin ;-) )
from now, i ll make a PR to update the man page (only master refers
ompi_unique)
Cheers,
Gilles
commit 7d2e3028d608163247975397a09f30dbe7bd192a
Author: Ralph Castain
Does the standard say anything about the default behavior? IIRC, we set it this
way because (a) we had no direction, and (b) it seemed just as reasonable as
the alternative (I believe we flipped a coin)
> On Jan 7, 2015, at 6:47 PM, Gilles Gouaillardet
> wrote:
Well, per the source code, this is not a bug but a feature :
from publish function from ompi/mca/pubsub/orte/pubsub_orte.c
ompi_info_get_bool(info, "ompi_unique", , );
if (0 == flag) {
/* uniqueness not specified - overwrite by default */
unique = false;
}
fwiw, and
Hmmm…I confess this API gets little, if any, testing as it is so seldom used,
so it is quite possible that a buglet has crept into it. I’ll take a look and
try to have something in 1.8.5.
Thanks!
Ralph
> On Jan 7, 2015, at 3:14 AM, Bernard Secher wrote:
>
> Hello,
>
>
Hello,
With the version openmpi-1.4.5 I got an error when I tried to publish
the same name twice with the MPI_Publish_name routine
With the version openmpi-1.8.4 I got no error when I published the same
name twice with the MPI_Publish_name routine
I used the attached script and source code