MPI\_ALLREDUCE\_INIT(3) Open MPI MPI\_ALLREDUCE\_INIT(3) MPI\_Allreduce <#mpi-allreduce>, MPI\_Iallreduce <#mpi-iallreduce>, MPI\_Allreduce\_init - Combines values from all processes and distributes the result back to all processes. # [SYNTAX](#SYNTAX) ## [C Syntax](#C_Syntax) ``` #include int MPI_Allreduce(const void *sendbuf, void *recvbuf, int count, MPI_Datatype datatype, MPI_Op op, MPI_Comm comm) int MPI_Iallreduce(const void *sendbuf, void *recvbuf, int count, MPI_Datatype datatype, MPI_Op op, MPI_Comm comm, MPI_Request *request) int MPI_Allreduce_init(const void *sendbuf, void *recvbuf, int count, MPI_Datatype datatype, MPI_Op op, MPI_Comm comm, MPI_Info info, MPI_Request *request) ``` ## [Fortran Syntax](#Fortran_Syntax) ``` USE MPI ! or the older form: INCLUDE 'mpif.h' MPI_ALLREDUCE(SENDBUF, RECVBUF, COUNT, DATATYPE, OP, COMM, IERROR) SENDBUF(*), RECVBUF(*) INTEGER COUNT, DATATYPE, OP, COMM, IERROR MPI_IALLREDUCE(SENDBUF, RECVBUF, COUNT, DATATYPE, OP, COMM, REQUEST, IERROR) SENDBUF(*), RECVBUF(*) INTEGER COUNT, DATATYPE, OP, COMM, REQUEST, IERROR MPI_ALLREDUCE_INIT(SENDBUF, RECVBUF, COUNT, DATATYPE, OP, COMM, INFO, REQUEST, IERROR) SENDBUF(*), RECVBUF(*) INTEGER COUNT, DATATYPE, OP, COMM, INFO, REQUEST, IERROR ``` ## [Fortran 2008 Syntax](#Fortran_2008_Syntax) ``` USE mpi_f08 MPI_Allreduce(sendbuf, recvbuf, count, datatype, op, comm, ierror) TYPE(*), DIMENSION(..), INTENT(IN) :: sendbuf TYPE(*), DIMENSION(..) :: recvbuf INTEGER, INTENT(IN) :: count TYPE(MPI_Datatype), INTENT(IN) :: datatype TYPE(MPI_Op), INTENT(IN) :: op TYPE(MPI_Comm), INTENT(IN) :: comm INTEGER, OPTIONAL, INTENT(OUT) :: ierror MPI_Iallreduce(sendbuf, recvbuf, count, datatype, op, comm, request, ierror) TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf INTEGER, INTENT(IN) :: count TYPE(MPI_Datatype), INTENT(IN) :: datatype TYPE(MPI_Op), INTENT(IN) :: op TYPE(MPI_Comm), INTENT(IN) :: comm TYPE(MPI_Request), INTENT(OUT) :: request INTEGER, OPTIONAL, INTENT(OUT) :: ierror MPI_Allreduce_init(sendbuf, recvbuf, count, datatype, op, comm, info, request, ierror) TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf INTEGER, INTENT(IN) :: count TYPE(MPI_Datatype), INTENT(IN) :: datatype TYPE(MPI_Op), INTENT(IN) :: op 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](#INPUT_PARAMETERS) - **sendbuf**: Starting address of send buffer (choice). - **count**: Number of elements in send buffer (integer). - **datatype**: Datatype of elements of send buffer (handle). - **op**: Operation (handle). - **comm**: Communicator (handle). - **info**: Info (handle, persistent only). # [OUTPUT PARAMETERS](#OUTPUT_PARAMETERS) - **recvbuf**: Starting address of receive buffer (choice). - **request**: Request (handle, non-blocking only). - **ierror**: Fortran only: Error status (integer). # [DESCRIPTION](#DESCRIPTION) Same as MPI\_Reduce <#mpi-reduce> except that the result appears in the receive buffer of all the group members. **Example 1:** A routine that computes the product of a vector and an array that are distributed across a group of processes and returns the answer at all nodes (compare with Example 2, with MPI\_Reduce <#mpi-reduce>, below). ``` SUBROUTINE PAR_BLAS2(m, n, a, b, c, comm) REAL a(m), b(m,n) ! local slice of array REAL c(n) ! result REAL sum(n) INTEGER n, comm, i, j, ierr ! local sum DO j= 1, n sum(j) = 0.0 DO i = 1, m sum(j) = sum(j) + a(i)*b(i,j) END DO END DO ! global sum CALL MPI_ALLREDUCE(sum, c, n, MPI_REAL, MPI_SUM, comm, ierr) ! return result at all nodes RETURN ``` **Example 2:** A routine that computes the product of a vector and an array that are distributed across a group of processes and returns the answer at node zero. ``` SUBROUTINE PAR_BLAS2(m, n, a, b, c, comm) REAL a(m), b(m,n) ! local slice of array REAL c(n) ! result REAL sum(n) INTEGER n, comm, i, j, ierr ! local sum DO j= 1, n sum(j) = 0.0 DO i = 1, m sum(j) = sum(j) + a(i)*b(i,j) END DO END DO ! global sum CALL MPI_REDUCE(sum, c, n, MPI_REAL, MPI_SUM, 0, comm, ierr) ! return result at node zero (and garbage at the other nodes) RETURN ``` # [USE OF IN-PLACE OPTION](#USE_OF_IN-PLACE_OPTION) When the communicator is an intracommunicator, you can perform an all-reduce operation in-place (the output buffer is used as the input buffer). Use the variable MPI\_IN\_PLACE as the value of *sendbuf* at all processes. Note that MPI\_IN\_PLACE is a special kind of value; it has the same restrictions on its use as MPI\_BOTTOM. Because the in-place option converts the receive buffer into a send-and-receive buffer, a Fortran binding that includes INTENT must mark these as INOUT, not OUT. # [WHEN COMMUNICATOR IS AN INTER-COMMUNICATOR](#WHEN_COMMUNICATOR_IS_AN_INTER-COMMUNICATOR) When the communicator is an inter-communicator, the reduce operation occurs in two phases. The data is reduced from all the members of the first group and received by all the members of the second group. Then the data is reduced from all the members of the second group and received by all the members of the first. The operation exhibits a symmetric, full-duplex behavior. When the communicator is an intra-communicator, these groups are the same, and the operation occurs in a single phase. # [NOTES ON COLLECTIVE OPERATIONS](#NOTES_ON_COLLECTIVE_OPERATIONS) The reduction functions ( MPI\_Op ) do not return an error value. As a result, if the functions detect an error, all they can do is either call MPI\_Abort <#mpi-abort> or silently skip the problem. Thus, if you change the error handler from MPI\_ERRORS\_ARE\_FATAL to something else, for example, MPI\_ERRORS\_RETURN , then no error may be indicated. # [ERRORS](#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>/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\_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 <#mpi-abort> was called on that communicator. If called on a window or file, acts as if MPI\_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. - **MPI\_ERRORS\_RETURN** Returns an error code to the application. MPI applications can also implement their own error handlers by calling: - MPI\_Comm\_create\_errhandler <#mpi-comm-create-errhandler> then MPI\_Comm\_set\_errhandler <#mpi-comm-set-errhandler> - MPI\_File\_create\_errhandler <#mpi-file-create-errhandler> then MPI\_File\_set\_errhandler <#mpi-file-set-errhandler> - MPI\_Session\_create\_errhandler <#mpi-session-create-errhandler> then MPI\_Session\_set\_errhandler <#mpi-session-set-errhandler> or at MPI\_Session\_init <#mpi-session-init> - MPI\_Win\_create\_errhandler <#mpi-win-create-errhandler> then MPI\_Win\_set\_errhandler <#mpi-win-set-errhandler> Note that MPI does not guarantee that an MPI program can continue past an error. 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. # [Copyright](#Copyright) 2003-2026, The Open MPI Community September 23, 2026 --- Document: MPI_Allreduce_init(3) from openmpi-cuda-12.9 5.0.11-r0 (Chainguard OS) Source: https://github.com/open-mpi/ompi