MPI_COMM_CREATE(3)
π MPI_COMM_CREATE π (Open MPI, section 3)
okay bestie, it's giving "make me a brand new communicator" and honestly? iconic. this is the MPI_COMM_CREATE(3) page from Open MPI, and we're gonna read it together like it's a group chat π
β¨ NAME β¨
MPI_Comm_create β creates a new communicator π«
πΈ SYNTAX πΈ
C Syntax
#include <mpi.h>
int MPI_Comm_create(MPI_Comm comm, MPI_Group group, MPI_Comm *newcomm)
Fortran Syntax
USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_COMM_CREATE(COMM, GROUP, NEWCOMM, IERROR)
INTEGER COMM, GROUP, NEWCOMM, IERROR
Fortran 2008 Syntax
USE mpi_f08
MPI_Comm_create(comm, group, newcomm, ierror)
TYPE(MPI_Comm), INTENT(IN) :: comm
TYPE(MPI_Group), INTENT(IN) :: group
TYPE(MPI_Comm), INTENT(OUT) :: newcomm
INTEGER, OPTIONAL, INTENT(OUT) :: ierror
π«Ά INPUT PARAMETER π«Ά
commβ the Communicator (a handle). your social circle, basically πgroupβ the Group (a handle), and it's gotta be a subset of the group ofcomm. no add-ons, no plus-ones, strictly a subgroup periodt.
π OUTPUT PARAMETERS π
newcommβ the new communicator (a handle). fresh out the oven β¨ierrorβ Fortran only: error status (integer). the lil receipt if something goes wrong π
π¦ DESCRIPTION π¦
So the way this works β this function creates a new communicator newcomm with a communication group defined by group and a new context. It sets newcomm to a new communicator that spans all the processes that are in the group. And for the processes that are not in the group? It sets newcomm to MPI_COMM_NULL. not invited, not in the group chat, blocked, no hard feelings π«
Every process has to call with a group argument that's a subgroup of the group tied to comm. And lowkey, that can even be MPI_GROUP_EMPTY. Different processes are allowed to pass different values for group β no cap.
BUT, and this is the important part bestie, if a process calls with a non-empty group, then all processes in that group must call the function with the same group as argument β like, the same processes in the same order. Otherwise the call is erroneous. π do not freestyle the ordering.
π₯ NOTES π₯
MPI_Comm_create is literally your plug for making a subset of processes for the purpose of separate MIMD computation, with a separate communication space. main character energy for real.
And here's the cute recursive part: newcomm, the one created by MPI_Comm_create, can be used in subsequent calls to MPI_Comm_create (or other communicator constructors) to further subdivide a computation into parallel sub-computations. it's giving subdivisions, it's giving nesting, it's giving β¨
If you want something more general, a more general service is provided by MPI_Comm_split <#mpi-comm-split>.
π ERRORS π
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 is returned, the current MPI error handler associated with the communication object (e.g., communicator, window, file) is called. If no communication object is associated with the MPI call, then the call is considered attached to MPI_COMM_SELF and will call the associated MPI error handler. When MPI_COMM_SELF is not initialized (i.e., before MPI_Init <#mpi-init> / MPI_Init_thread <#mpi-init-thread>, after MPI_Finalize <#mpi-finalize>, or when using the Sessions Model exclusively) the error raises the initial error handler. The initial error handler can be changed by calling MPI_Comm_set_errhandler <#mpi-comm-set-errhandler> on MPI_COMM_SELF when using the World model, or the mpi_initial_errhandler CLI argument to mpiexec or info key to MPI_Comm_spawn <#mpi-comm-spawn> / MPI_Comm_spawn_multiple <#mpi-comm-spawn-multiple>. If no other appropriate error handler has been set, then the MPI_ERRORS_RETURN error handler is called for MPI I/O functions and the MPI_ERRORS_ABORT error handler is called for all other MPI functions.
Open MPI includes three predefined error handlers that can be used:
MPI_ERRORS_ARE_FATALβ causes the program to abort all connected MPI processes. π no survivors.MPI_ERRORS_ABORTβ an error handler that can be invoked on a communicator, window, file, or session. When called on a communicator, it acts as ifMPI_Abort <#mpi-abort>was called on that communicator. If called on a window or file, acts as ifMPI_Abort <#mpi-abort>was called on a communicator containing the group of processes in the corresponding window or file. If called on a session, aborts only the local process. βοΈ it knows exactly how much to cut.MPI_ERRORS_RETURNβ returns an error code to the application. the polite one π«Ά
MPI applications can also implement their own error handlers by calling:
MPI_Comm_create_errhandler <#mpi-comm-create-errhandler>thenMPI_Comm_set_errhandler <#mpi-comm-set-errhandler>MPI_File_create_errhandler <#mpi-file-create-errhandler>thenMPI_File_set_errhandler <#mpi-file-set-errhandler>MPI_Session_create_errhandler <#mpi-session-create-errhandler>thenMPI_Session_set_errhandler <#mpi-session-set-errhandler>or atMPI_Session_init <#mpi-session-init>MPI_Win_create_errhandler <#mpi-win-create-errhandler>thenMPI_Win_set_errhandler <#mpi-win-set-errhandler>
Note that MPI does not guarantee that an MPI program can continue past an error. like, it might just be over bestie π
See the MPI man page <#open-mpi> for a full list of MPI error codes <#open-mpi-errors>.
See the Error Handling section of the MPI-3.1 standard for more information.
See also:
MPI_Comm_split <#mpi-comm-split>
π Copyright π
Copyright 2003-2026, The Open MPI Community
and that's the whole page bestie, you ate it up fr fr πβ¨ this was the Open MPI MPI_COMM_CREATE(3) page, dated Sep 10, 2026, brought to you by the Open MPI Community (2003-2026) β go forth and subdivide your communicators, just keep the group order exactly the same or it's game over π
π¦
π¦ where this page comes from
- ubuntu stonking: openmpi-doc 5.0.10-6 Β· other
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms