MPI_WAITSOME(3)

✨ MPI_Waitsome ✨

okay bestie, welcome to the MPI_WAITSOME(3) page from the Open MPI manual πŸ’– this is the one where you sit around waiting for some of your communications to finish β€” not all of them, not one specific one, just… some. it's giving "I'll take whoever's ready first" energy and honestly? iconic.


🌸 SYNOPSIS

C Syntax

#include <mpi.h>

int MPI_Waitsome(int incount, MPI_Request array_of_requests[],
     int *outcount, int array_of_indices[],
     MPI_Status array_of_statuses[])

Fortran Syntax

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_WAITSOME(INCOUNT, ARRAY_OF_REQUESTS, OUTCOUNT,
             ARRAY_OF_INDICES, ARRAY_OF_STATUSES, IERROR)
     INTEGER INCOUNT, ARRAY_OF_REQUESTS(*), OUTCOUNT
     INTEGER ARRAY_OF_INDICES(*)
     INTEGER ARRAY_OF_STATUSES(MPI_STATUS_SIZE*)
     INTEGER IERROR

Fortran 2008 Syntax

USE mpi_f08
MPI_Waitsome(incount, array_of_requests, outcount, array_of_indices,
             array_of_statuses, ierror)
     INTEGER, INTENT(IN) :: incount
     TYPE(MPI_Request), INTENT(INOUT) :: array_of_requests(incount)
     INTEGER, INTENT(OUT) :: outcount, array_of_indices(*)
     TYPE(MPI_Status) :: array_of_statuses(*)
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

πŸ’Œ INPUT PARAMETERS

  • incount β€” the length of array_of_requests, an integer. it's literally just "how many things are in my list" πŸ’…
  • array_of_requests β€” the array of requests, which is an array of handles. your little pile of pending ops πŸŽ€

🫢 OUTPUT PARAMETERS

  • outcount β€” how many requests completed, an integer. the receipts, bestie ✨
  • array_of_indices β€” an array of integers holding the indices of the operations that completed πŸ‘‘
  • array_of_statuses β€” an array of status objects for the operations that completed 🌸
  • ierror β€” Fortran only: the error status, an integer.

πŸ¦‹ DESCRIPTION

so here's the vibe: MPI_Waitsome waits until at least one of the operations associated with the active handles in your list has completed. that's the whole bit. it will not wait for all of them, and it will not wait for a specific one β€” it just needs somebody to finish πŸ₯°

Then it reports back:

  • in outcount, it returns the number of requests from array_of_requests that have completed.
  • in the first outcount locations of array_of_indices, it returns the indices of those operations β€” index within the array array_of_requests. heads up: that array is indexed from 0 in C and from 1 in Fortran. no cap, that trips people up every single time 😭
  • in the first outcount locations of the array array_of_status, it returns the status for those completed operations.

If a request that completed was allocated by a nonblocking communication call, then it is deallocated, and the associated handle is set to MPI_REQUEST_NULL. clean up after yourself, girlboss πŸ’…

If the list contains no active handles, then the call returns immediately with outcount = MPI_UNDEFINED. not me showing up to an empty party and immediately leaving πŸ™ˆ

When one or more of the communications completed by MPI_Waitsome fails, then it is desirable to return specific information on each communication. The arguments outcount, array_of_indices, and array_of_statuses will be adjusted to indicate completion of all communications that have succeeded or failed. The call will return the error code MPI_ERR_IN_STATUS, and the error field of each status returned will be set to indicate success or to indicate the specific error that occurred. The call will return MPI_SUCCESS if no request resulted in an error, and will return another error code if it failed for other reasons (such as invalid arguments). In such cases, it will not update the error fields of the statuses.

If your application does not need to examine the array_of_statuses field, you can save resources by using the predefined constant MPI_STATUSES_IGNORE β€” you can use it as a special value for the array_of_statuses argument. skipping the homework, we love to see it ✨

Example: Same code as the example in the MPI_Waitany man page, but using MPI_Waitsome.

CALL MPI_COMM_SIZE(comm, size, ierr)
CALL MPI_COMM_RANK(comm, rank, ierr)
IF(rank > 0) THEN         ! client code
    DO
       CALL MPI_ISEND(a, n, MPI_REAL, 0, tag, comm, request, ierr)
       CALL MPI_WAIT(request, status, ierr)
    END DO
ELSE         ! rank=0: server code
    DO i=1, size-1
       CALL MPI_IRECV(a(1,i), n, MPI_REAL, i, tag, &
                      comm, requests(i), ierr)
    END DO
    DO
       CALL MPI_WAITSOME(size, request_list, numdone, &
                        indices, statuses, ierr)
       DO i=1, numdone
          CALL DO_SERVICE(a(1, indices(i)))
          CALL MPI_IRECV(a(1, indices(i)), n, MPI_REAL, i, tag, &
                       comm, requests(indices(i)), ierr)
       END DO
    END DO
END IF

🌈 NOTES

The array of indices are in the range 0 to incount-1 for C and in the range 1 to incount for Fortran. yes, again. remember it this time bestie πŸŽ€


πŸ’€ 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 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_thread, after 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 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_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_FATAL β€” Causes the program to abort all connected MPI processes. πŸ’€ she's dramatic and she means it
  • MPI_ERRORS_ABORT β€” 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, 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, aborts only the local process.
  • MPI_ERRORS_RETURN β€” Returns an error code to the application. the polite one, she just tells you what went wrong πŸ’Œ

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

Note that MPI does not guarantee that an MPI program can continue past an error. 😭 fr fr, once you're in error territory there's no promise you're coming back.

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.

For each invocation of MPI_Waitsome, if one or more requests generate an MPI error, only the first MPI request that caused an error will be passed to its corresponding error handler. No other error handlers will be invoked (even if multiple requests generated errors). However, all requests that generate an error will have a relevant error code set in the corresponding status.MPI_ERROR field (unless MPI_STATUSES_IGNORE was used). so like, one handler gets the tea, but everyone still gets their error code noted πŸ“

If the invoked error handler allows MPI_Waitsome to return to the caller, the value MPI_ERR_IN_STATUS will be returned in the C and Fortran bindings.

See also:

  • MPI_Comm_set_errhandler
  • MPI_File_set_errhandler
  • MPI_Test
  • MPI_Testall
  • MPI_Testany
  • MPI_Testsome
  • MPI_Wait
  • MPI_Waitall
  • MPI_Waitany
  • MPI_Win_set_errhandler

2003-2026, The Open MPI Community


aaaand that's the whole page bestie, you ate that πŸ’– go forth and wait on some of your requests like the main character you are. this was MPI_WAITSOME(3) from the Open MPI crew, dated Sep 10, 2026 β€” mwah, no notes, periodt βœ¨πŸ¦‹πŸŽ€

πŸ“¦ where this page comes from