[Mpi3-hybridpm] INIT/FINALIZE, with corrections suggested by Darius
dinan at mcs.anl.gov
Mon Feb 6 11:09:57 CST 2012
Thanks for the feedback. I created a ticket to track this:
On 2/5/12 7:09 PM, Martin Schulz wrote:
> Hi Jim,
> Just as a comment: if you want to go down this route, it may be worth to think about a different approach to describing this list, since it is starting to get large. In addition to the three routines mentioned in 2.2, MPI_Get_library_version and all routines starting with MPI_T_ will be allowed as well in MPI 3 (based on the second votes for tickets #204 and #266 at the last forum).
> On Feb 3, 2012, at 3:17 PM, Jim Dinan wrote:
>> Right, that change wouldn't go in this ticket. My question was more
>> along the lines of: Is this something worthwhile in the broader MPI
>> sense? If so, I can create a ticket so we don't lose it.
>> On 2/3/12 2:48 PM, Pavan Balaji wrote:
>>> On 02/03/2012 02:36 PM, Jim Dinan wrote:
>>>> As long as we're updating this text, do we want to allow any additional
>>>> routines before MPI_Init/after MPI_Finalize (e.g. MPI_Wtime)?
>>> Why? The changes made so far are all addressing the limitations that
>>> need to be addressed for the endpoints proposal. They are not being made
>>> lightly. Adding MPI_WTIME is irrelevant for this.
>>> That could be a separate ticket if you like. But there's no reason to
>>> mix these two together.
>>> -- Pavan
>> Mpi3-hybridpm mailing list
>> Mpi3-hybridpm at lists.mpi-forum.org
> Martin Schulz, schulzm at llnl.gov, http://people.llnl.gov/schulzm
> CASC @ Lawrence Livermore National Laboratory, Livermore, USA
> Mpi3-hybridpm mailing list
> Mpi3-hybridpm at lists.mpi-forum.org
More information about the mpiwg-hybridpm