MPI_INEIGHBOR_ALLGATHER(3)

📦
version 5.0.10 · 4 versions
, in ubuntu stonking

💅 MPI_INEIGHBOR_ALLGATHER 💅 (the Open MPI one, bestie)

okay so first thing — this page is a whole family of routines and they all live under the <rainbow>Open MPI</rainbow> umbrella. You got MPI_Neighbor_allgather, you got MPI_Ineighbor_allgather, and you got MPI_Neighbor_allgather_init (yes the header says MPI_NEIGHBOR_ALLGATHER like three times, it's giving ✨one big happy routine✨).

What do they do? Gathers and distributes data from and to all neighbors. That's it. That's the whole vibe. 🦋

✨ SYNTAX ✨

🎀 C Syntax

#include <mpi.h>

int MPI_Neighbor_allgather(const void *sendbuf, int  sendcount,
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
      MPI_Datatype recvtype, MPI_Comm comm)

int MPI_Ineighbor_allgather(const void *sendbuf, int  sendcount,
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
      MPI_Datatype recvtype, MPI_Comm comm, MPI_Request req)

int MPI_Neighbor_allgather_init(const void *sendbuf, int  sendcount,
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
      MPI_Datatype recvtype, MPI_Comm comm, MPI_Infoinfo, MPI_Request req)

(and yes, copy it character for character, we do not "fix" anything in this house 💀)

🧚 Fortran Syntax

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_NEIGHBOR_ALLGATHER(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNT,
             RECVTYPE, COMM, IERROR)
     <type>  SENDBUF (*), RECVBUF (*)
     INTEGER SENDCOUNT, SENDTYPE, RECVCOUNT, RECVTYPE, COMM,
     INTEGER IERROR

MPI_INEIGHBOR_ALLGATHER(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNT,
             RECVTYPE, COMM, REQUEST, IERROR)
     <type>  SENDBUF (*), RECVBUF (*)
     INTEGER SENDCOUNT, SENDTYPE, RECVCOUNT, RECVTYPE, COMM,
     INTEGER REQUEST, IERROR

MPI_NEIGHBOR_ALLGATHER_INIT(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNT,
             RECVTYPE, COMM, INFO, IREQUEST, IERROR)
     <type>  SENDBUF (*), RECVBUF (*)
     INTEGER SENDCOUNT, SENDTYPE, RECVCOUNT, RECVTYPE, COMM,
     INTEGER INFO, REQUEST, IERROR

🌸 Fortran 2008 Syntax

USE mpi_f08
MPI_Neighbor_allgather(sendbuf, sendcount, sendtype, recvbuf, recvcount,
             recvtype, comm, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN) :: sendbuf
     TYPE(*), DIMENSION(..) :: recvbuf
     INTEGER, INTENT(IN) :: sendcount, recvcount
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Ineighbor_allgather(sendbuf, sendcount, sendtype, recvbuf, recvcount,
             recvtype, comm, request, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: sendcount, recvcount
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     TYPE(MPI_Request), INTENT(OUT) :: request
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Neighbor_allgather_init(sendbuf, sendcount, sendtype, recvbuf, recvcount,
             recvtype, comm, info, request, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: sendcount, recvcount
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     TYPE(MPI_Info), INTENT(IN) :: info
     TYPE(MPI_Request), INTENT(OUT) :: request
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

🍓 INPUT PARAMETERS 🍓

  • sendbuf — the starting address of the send buffer (a choice). where ur stuff is coming from 💌
  • sendcount — how many elements live in the send buffer (integer).
  • sendtype — the datatype of the send buffer elements (a handle).
  • recvbuf — the starting address of the recv buffer (a choice).
  • recvcount — number of elements received from any process (integer).
  • recvtype — datatype of the receive buffer elements (a handle).
  • comm — the communicator (a handle).

🫶 OUTPUT PARAMETERS 🫶

  • recvbuf — address of the receive buffer (a choice). yes bestie, it shows up twice, it's on the input list and the output list, that's just how it is 💖
  • request — the request (a handle, non-blocking only). so like, only for the I version, ok?
  • ierror — Fortran only: the error status (integer).

💜 DESCRIPTION 💜

So MPI_Neighbor_allgather is literally MPI_Allgather's cousin, but ✨*selective*✨ — the difference is that only the neighboring processes get the result, instead of everybody. Main character energy but with a small guest list. The neighbors and the buffer layout are decided by the topology of comm.

Here's the important bit, no cap: the type signature tied to sendcount, sendtype at one process must be equal to the type signature tied to recvcount, recvtype at any other process. Same signature, every time. 📏

🗺️ NEIGHBOR ORDERING 🗺️

okay this is the part where the ordering rules show up, and yes they're different depending on which topology u built. pay attention bestie:

  • For a distributed graph topology, made with MPI_Dist_graph_create, the sequence of neighbors in the send and receive buffers at each process is defined as the sequence returned by MPI_Dist_graph_neighbors for destinations and sources, respectively.
  • For a general graph topology, made with MPI_Graph_create, the order of neighbors in the send and receive buffers is defined as the sequence of neighbors returned by MPI_Graph_neighbors. also, gentle heads up: general graph topologies should generally be replaced by the distributed graph topologies. like, lowkey deprecated vibes 👀
  • For a Cartesian topology, made with MPI_Cart_create, the sequence of neighbors in the send and receive buffers at each process is defined by the order of the dimensions — first the neighbor in the negative direction, then in the positive direction with displacement 1. The numbers of sources and destinations in the communication routines are 2*ndims, with ndims defined in MPI_Cart_create. And if a neighbor doesn't exist — like, say, at the border of a Cartesian topology on a non-periodic virtual grid dimension (i.e., periods[…]==false) — then that neighbor is defined to be MPI_PROC_NULL.

And if a neighbor in any of the functions is MPI_PROC_NULL, then the neighborhood collective communication behaves like a point-to-point communication with MPI_PROC_NULL in that direction. Translation: the buffer is still part of the sequence of neighbors, but it is neither communicated nor updated. it's just… there. standing in the group chat not saying anything 🧍

⚠️ NOTES ⚠️

The MPI_IN_PLACE option for sendbuf is not meaningful for this operation. don't do it, it does nothing for u here 💀

😭 ERRORS 😭

Almost all MPI routines return an error value — C routines as the return result of the function, and Fortran routines in the last argument.

Before the error value gets returned, the current MPI error handler associated with the communication object (communicator, window, file, whatever) is called. If there's no communication object associated with the MPI call, then the call counts as attached to MPI_COMM_SELF and will call the associated MPI error handler. When MPI_COMM_SELF isn't initialized (i.e., before MPI_Init/MPI_Init_thread, after MPI_Finalize, or when using the Sessions Model exclusively) the error raises the initial error handler. That initial error handler can be changed by calling MPI_Comm_set_errhandler on MPI_COMM_SELF when using the World model, or by using the mpi_initial_errhandler CLI argument to mpiexec, or the info key to MPI_Comm_spawn/MPI_Comm_spawn_multiple. If no other appropriate error handler has been set, then MPI_ERRORS_RETURN is called for MPI I/O functions and MPI_ERRORS_ABORT is called for all other MPI functions.

Open MPI ships three predefined error handlers u can use:

  • MPI_ERRORS_ARE_FATAL — Causes the program to abort all connected MPI processes. 💀
  • MPI_ERRORS_ABORT — An error handler that can be invoked on a communicator, window, file, or session. When called on a communicator, it acts as if MPI_Abort was called on that communicator. If called on a window or file, it acts as if MPI_Abort was called on a communicator containing the group of processes in the corresponding window or file. If called on a session, it aborts only the local process.
  • MPI_ERRORS_RETURN — Returns an error code to the application. polite. ✨

MPI applications can also implement their own error handlers by calling:

  • MPI_Comm_create_errhandler then MPI_Comm_set_errhandler
  • MPI_File_create_errhandler then MPI_File_set_errhandler
  • MPI_Session_create_errhandler then MPI_Session_set_errhandler or at MPI_Session_init
  • MPI_Win_create_errhandler then MPI_Win_set_errhandler
bestie, listen: MPI does not guarantee that an MPI program can continue past an error. it might just… not. 🫠

See the MPI man page for a full list of MPI error codes.

See the Error Handling section of the MPI-3.1 standard for more information.

See also:

  • MPI_Neighbor_allgatherv
  • MPI_Cart_create
  • MPI_Garph_create
  • MPI_Dist_graph_create
  • MPI_Gather

2003-2026, The Open MPI Community

and that's the whole page, periodt 💅✨ no cap, u now know everything there is to know about this routine from the Open MPI crew, dated Sep 10, 2026 — go forth and neighbor-allgather, bestie 🦋🎀💖

📦 where this page comes from