Big MPI--Apply Big MPI -- Large Count / Large Displacement Ideas to the Topologies Chapter

[tonyskjellum]

Problem

See Motivation with Ticket #80.

Proposal

See Proposal with Ticket #80. [As may be modified by discussion in Barcelona.]

Changes to the Text

Add _x function APIs where applicable.

Impact on Implementations

TBD.

Impact on Users

New APIs that support MPI_Count / MPI_Aint (where applicable) are done in analogy to Chapter 5.

References

Ticket #80.

PR https://github.com/mpi-forum/mpi-standard/pull/56

[tonyskjellum]

Here is the zeroth version of this chapter (see topologies chapter only).

mpi32-report-ticket105-19sep18.pdf

[RolfRabenseifner]

Dear Tony, Review: - MPI_Neighbor_allgather : all _x are missing - int MPI_Neighbor_allgatherv_x(const void* sendbuf, MPI_Count sendcount, MPI_Datatype sendtype, void* recvbuf, const_MPI_Count recvcounts[], const_MPI_Count --> const MPI_Count - MPI_NEIGHBOR_ALLGATHERV_X(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNTS, DISPLS, RECVTYPE, COMM, IERROR) <type> SENDBUF(*), RECVBUF(*) INTEGER SENDCOUNT, SENDTYPE, RECVCOUNTS(*), DISPLS(*), RECVTYPE, COMM, IERROR You forgot to make them MPI_Count - MPI_NEIGHBOR_ALLTOALL_X(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNT, ... You forgot to make them MPI_Count - MPI_NEIGHBOR_ALLTOALLV_X(SENDBUF, SENDCOUNTS, SDISPLS, SENDTYPE, RECVBUF, RECVCOUNTS, RDISPLS, You forgot to make them MPI_Count - MPI_Neighbor_alltoallw_x C: completetly missing F_90: correct Old Fortran: You forgot to make them MPI_Count and you still have old Name without _x Best regards Rolf

…

----- Original Message -----

From: "github notifications" ***@***.***> To: "mpi-forum" ***@***.***> Cc: "Subscribed" ***@***.***> Sent: Wednesday, September 19, 2018 6:12:30 PM Subject: Re: [mpi-forum/mpi-issues] Big MPI--Apply Big MPI -- Large Count / Large Displacement Ideas to the Topologies Chapter (#105) Here is the zeroth version of this chapter (see topologies chapter only). [mpi32-report-ticket105-19sep18.pdf](https://github.com/mpi-forum/mpi-issues/files/2397843/mpi32-report-ticket105-19sep18.pdf) -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: #105 (comment)

-- Dr. Rolf Rabenseifner . . . . . . . . . .. email [email protected] . High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 . University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 . Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner . Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) .

[tonyskjellum]

Latest update:

mpi32-report-ticket105-19sep18-2333.pdf

[tonyskjellum]

Latest update (improves change log):

mpi32-report-ticket105-20sep18-1044.pdf

[tonyskjellum]

An update is forthcoming that adds _X APIs for the persistent API approved with Ticket #25 at the Barcelona meeting.

[wesbland]

@tonyskjellum / @puribangalore - Is this issue replaced by #137? Can we close this?

[puribangalore]

Yes, we can close this issue.

…

On Wed, Oct 7, 2020 at 10:33 AM Wesley Bland ***@***.***> wrote: @tonyskjellum <https://github.com/tonyskjellum> / @puribangalore <https://github.com/puribangalore> - Is this issue replaced by #137 <#137>? Can we close this? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#105 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AD2CIBIAOMKYSGJFHPISVPLSJSC3LANCNFSM4FV7N7OQ> .