George,
please disregard my previous comment.
new_comm is passed to ompi_comm_enable (and *not* &new_comm)
so you are right, new_comm cannot be MPI_COMM_NULL at topo_base_create.c:172
that being said, ompi_comm_enable invokes ompi_comm_activate that
OBJ_RELEASE
new_comm if an error occurs, so th
George,
In topo_base_create.c at line 171, ompi_comm_enable ends up calling
ompi_comm_activate which assigns MPI_COMM_NULL to new_comm when an error
occurs
( see bail_on_error label )
So unless the correct fix is to change the behavior of ompi_comm_activate,
I think the change is correct.
Makes
Die batman Lampe, ein toller Einfall!
!
2015-03-09 10:11 GMT-06:00 Mike Dubman :
>
> Hello,
>
> Please check updated OMPI wiki page for detailed information for Jenkins
> testing of OMPI repositories.
>
> https://github.com/open-mpi/ompi/wiki/PRJenkins
>
> Comments and suggestions are welcome.
>
Hello,
Please check updated OMPI wiki page for detailed information for Jenkins
testing of OMPI repositories.
https://github.com/open-mpi/ompi/wiki/PRJenkins
Comments and suggestions are welcome.
--
Kind Regards,
M.
Gilles,
I might misread these commit, but the changes proposed here do not look correct
to me. At no moment the new_comm can be equal to MPI_COMM_NULL in this code
(especially not at line 172 in the too_base_cart_create.c).
George.
> On Mar 9, 2015, at 02:26 , [email protected] wrote:
>
>
