Okay, so let's revert this commit and instead CMR over the one that includes the required code.

On Feb 6, 2014, at 9:16 AM, Mike Dubman <miked@dev.mellanox.co.il> wrote:

It seems that similar code in not in v1.7 tree.


On Thu, Feb 6, 2014 at 2:40 PM, George Bosilca <bosilca@icl.utk.edu> wrote:
This commit is unnecessary. The call to delete_proc is already there, few lines above your own patch. It was introduced on Jan 26 2014 with commit https://svn.open-mpi.org/trac/ompi/changeset/30430.

  George.



On Feb 6, 2014, at 09:38 , svn-commit-mailer@open-mpi.org wrote:

> Author: miked (Mike Dubman)
> Date: 2014-02-06 03:38:32 EST (Thu, 06 Feb 2014)
> New Revision: 30571
> URL: https://svn.open-mpi.org/trac/ompi/changeset/30571
>
> Log:
> OMPI: add call to del_procs
>
> fixed by AlexM, reviewed by miked
> cmr=v1.7.5:reviewer=ompi-rm1.7
>
> Text files modified:
>   trunk/ompi/runtime/ompi_mpi_finalize.c |    15 +++++++++++++++
>   1 files changed, 15 insertions(+), 0 deletions(-)
>
> Modified: trunk/ompi/runtime/ompi_mpi_finalize.c
> ==============================================================================
> --- trunk/ompi/runtime/ompi_mpi_finalize.c    Wed Feb  5 17:49:26 2014        (r30570)
> +++ trunk/ompi/runtime/ompi_mpi_finalize.c    2014-02-06 03:38:32 EST (Thu, 06 Feb 2014)      (r30571)
> @@ -94,6 +94,9 @@
>     opal_list_item_t *item;
>     struct timeval ompistart, ompistop;
>     ompi_rte_collective_t *coll;
> +    ompi_proc_t** procs;
> +    size_t nprocs;
> +
>
>     /* Be a bit social if an erroneous program calls MPI_FINALIZE in
>        two different threads, otherwise we may deadlock in
> @@ -150,6 +153,18 @@
>        MPI lifetime, to get better latency when not using TCP */
>     opal_progress_event_users_increment();
>
> +
> +    if (NULL == (procs = ompi_proc_world(&nprocs))) {
> +        return OMPI_ERROR;
> +    }
> +
> +    if (OMPI_SUCCESS != (ret = MCA_PML_CALL(del_procs(procs, nprocs)))) {
> +        free(procs);
> +        return ret;
> +    }
> +    free(procs);
> +
> +
>     /* check to see if we want timing information */
>     if (ompi_enable_timing != 0 && 0 == OMPI_PROC_MY_NAME->vpid) {
>         gettimeofday(&ompistart, NULL);
> _______________________________________________
> svn mailing list
> svn@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/svn

_______________________________________________
devel mailing list
devel@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel

_______________________________________________
devel mailing list
devel@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel