Skip to content

Latest commit

 

History

History
114 lines (79 loc) · 3.15 KB

MPI_Startall.3.rst

File metadata and controls

114 lines (79 loc) · 3.15 KB

MPI_Startall

:ref:`MPI_Startall` - Starts a collection of requests.

SYNTAX

C Syntax

#include <mpi.h>

int MPI_Startall(int count, MPI_Request array_of_requests[])

Fortran Syntax

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_STARTALL(COUNT, ARRAY_OF_REQUESTS, IERROR)
     INTEGER COUNT, ARRAY_OF_REQUESTS(*), IERROR

Fortran 2008 Syntax

USE mpi_f08
MPI_Startall(count, array_of_requests, ierror)
     INTEGER, INTENT(IN) :: count
     TYPE(MPI_Request), INTENT(INOUT) :: array_of_requests(count)
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

INPUT PARAMETER

  • count: List length (integer).

INPUT/OUTPUT PARAMETER

  • array_of_requests: Array of requests (array of handle).

OUTPUT PARAMETER

  • ierror: Fortran only: Error status (integer).

DESCRIPTION

Starts all communications associated with requests in array_of_requests. A call to MPI_Startall(count, array_of_requests) has the same effect as calls to :ref:`MPI_Start` (&array_of_requests[i]), executed for i=0 ,..., count-1, in some arbitrary order.

A communication started with a call to :ref:`MPI_Start` or :ref:`MPI_Startall` is completed by a call to :ref:`MPI_Wait`, :ref:`MPI_Test`, or one of the derived functions :ref:`MPI_Waitany`, :ref:`MPI_Testany`, :ref:`MPI_Waitall`, :ref:`MPI_Testall`, :ref:`MPI_Waitsome`, :ref:`MPI_Testsome` (these are described in Section 3.7.5 of the MPI-1 Standard, "Multiple Completions"). The request becomes inactive after successful completion by such a call. The request is not deallocated, and it can be activated anew by another :ref:`MPI_Start` or :ref:`MPI_Startall` call.

A persistent request is deallocated by a call to :ref:`MPI_Request_free` (see Section 3.7.3 of the MPI-1 Standard, "Communication Completion").

The call to :ref:`MPI_Request_free` can occur at any point in the program after the persistent request was created. However, the request will be deallocated only after it becomes inactive. Active receive requests should not be freed. Otherwise, it will not be possible to check that the receive has completed. It is preferable, in general, to free requests when they are inactive. If this rule is followed, then the persistent communication request functions will be invoked in a sequence of the form,
Create (Start Complete)* Free

where * indicates zero or more repetitions. If the same communication object is used in several concurrent threads, it is the user's responsibility to coordinate calls so that the correct sequence is obeyed.

A send operation initiated with :ref:`MPI_Start` can be matched with any receive operation and, likewise, a receive operation initiated with :ref:`MPI_Start` can receive messages generated by any send operation.

ERRORS

.. seealso::
   * :ref:`MPI_Bsend_init`
   * :ref:`MPI_Rsend_init`
   * :ref:`MPI_Send_init`
   * :ref:`MPI_Ssend_init`
   * :ref:`MPI_Recv_init`
   * :ref:`MPI_Start`
   * :ref:`MPI_Request_free`