MPI\_PROBE(3) Open MPI MPI\_PROBE(3)

MPI\_Probe — Blocking test for a message.

# [SYNTAX](#SYNTAX)

## [C Syntax](#C_Syntax)

```
#include <mpi.h>
int MPI_Probe(int source, int tag, MPI_Comm comm, MPI_Status *status)
```

## [Fortran Syntax](#Fortran_Syntax)

```
USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_PROBE(SOURCE, TAG, COMM, STATUS, IERROR)


     INTEGER SOURCE, TAG, COMM, STATUS(MPI_STATUS_SIZE), IERROR
```

## [Fortran 2008 Syntax](#Fortran_2008_Syntax)

```
USE mpi_f08
MPI_Probe(source, tag, comm, status, ierror)


     INTEGER, INTENT(IN) :: source, tag


     TYPE(MPI_Comm), INTENT(IN) :: comm


     TYPE(MPI_Status) :: status


     INTEGER, OPTIONAL, INTENT(OUT) :: ierror
```

# [INPUT PARAMETERS](#INPUT_PARAMETERS)

- **source**: Source rank or **MPI\_ANY\_SOURCE** (integer).
- **tag**: Tag value or **MPI\_ANY\_TAG** (integer).
- **comm**: Communicator (handle).

# [OUTPUT PARAMETERS](#OUTPUT_PARAMETERS)

- **status**: Status object (status).
- **ierror**: Fortran only: Error status (integer).

# [DESCRIPTION](#DESCRIPTION)

The MPI\_Probe and MPI\_Iprobe &lt;#mpi-iprobe&gt; operations allow checking of incoming messages, without actual receipt of them. The user can then decide how to receive them, based on the information returned by the probe in the status variable. For example, the user may allocate memory for the receive buffer, according to the length of the probed message.

MPI\_Probe behaves like MPI\_Iprobe &lt;#mpi-iprobe&gt; except that it is a blocking call that returns only after a matching message has been found.

If your application does not need to examine the *status* field, you can save resources by using the predefined constant **MPI\_STATUS\_IGNORE** as a special value for the *status* argument.

The semantics of MPI\_Probe and MPI\_Iprobe &lt;#mpi-iprobe&gt; guarantee progress: If a call to MPI\_Probe has been issued by a process, and a send that matches the probe has been initiated by some process, then the call to MPI\_Probe will return, unless the message is received by another concurrent receive operation (that is executed by another thread at the probing process). Similarly, if a process busy waits with MPI\_Iprobe &lt;#mpi-iprobe&gt; and a matching message has been issued, then the call to MPI\_Iprobe &lt;#mpi-iprobe&gt; will eventually return *flag* = true unless the message is received by another concurrent receive operation.

**Example 1:** Use blocking probe to wait for an incoming message.

```
CALL MPI_COMM_RANK(comm, rank, ierr)
IF (rank == 0) THEN


   CALL MPI_SEND(i, 1, MPI_INTEGER, 2, 0, comm, ierr)
ELSE IF(rank == 1) THEN


   CALL MPI_SEND(x, 1, MPI_REAL, 2, 0, comm, ierr)
ELSE   ! rank == 2


   DO i=1, 2


      CALL MPI_PROBE(MPI_ANY_SOURCE, 0,


                     comm, status, ierr)


      IF (status(MPI_SOURCE) = 0) THEN


         CALL MPI_RECV(i, 1, MPI_INTEGER, 0, 0, status, ierr)


      ELSE


         CALL MPI_RECV(x, 1, MPI_REAL, 1, 0, status, ierr)


      END IF


   END DO
END IF
```

Each message is received with the right type.

**Example 2:** A program similar to the previous example, but with a problem.

```
CALL MPI_COMM_RANK(comm, rank, ierr)
IF (rank == 0) THEN


   CALL MPI_SEND(i, 1, MPI_INTEGER, 2, 0, comm, ierr)
ELSE IF(rank == 1) THEN


   CALL MPI_SEND(x, 1, MPI_REAL, 2, 0, comm, ierr)
ELSE


   DO i=1, 2


      CALL MPI_PROBE(MPI_ANY_SOURCE, 0,


                     comm, status, ierr)


      IF (status(MPI_SOURCE) == 0) THEN


         CALL MPI_RECV(i, 1, MPI_INTEGER, MPI_ANY_SOURCE, &


                       0, status, ierr)


      ELSE


         CALL MPI_RECV(x, 1, MPI_REAL, MPI_ANY_SOURCE, &


                       0, status, ierr)


      END IF


   END DO
END IF
```

We slightly modified Example 2, using **MPI\_ANY\_SOURCE** as the source argument in the two receive calls in statements labeled 100 and 200. The program is now incorrect: The receive operation may receive a message that is distinct from the message probed by the preceding call to MPI\_Probe.

# [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 &lt;#mpi-init&gt;/MPI\_Init\_thread &lt;#mpi-init-thread&gt;, after MPI\_Finalize &lt;#mpi-finalize&gt;, 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 &lt;#mpi-comm-set-errhandler&gt; 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 &lt;#mpi-comm-spawn&gt;/MPI\_Comm\_spawn\_multiple &lt;#mpi-comm-spawn-multiple&gt;. 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 &lt;#mpi-abort&gt; was called on that communicator. If called on a window or file, acts as if MPI\_Abort &lt;#mpi-abort&gt; 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 &lt;#mpi-comm-create-errhandler&gt; then MPI\_Comm\_set\_errhandler &lt;#mpi-comm-set-errhandler&gt;
- MPI\_File\_create\_errhandler &lt;#mpi-file-create-errhandler&gt; then MPI\_File\_set\_errhandler &lt;#mpi-file-set-errhandler&gt;
- MPI\_Session\_create\_errhandler &lt;#mpi-session-create-errhandler&gt; then MPI\_Session\_set\_errhandler &lt;#mpi-session-set-errhandler&gt; or at MPI\_Session\_init &lt;#mpi-session-init&gt;
- MPI\_Win\_create\_errhandler &lt;#mpi-win-create-errhandler&gt; then MPI\_Win\_set\_errhandler &lt;#mpi-win-set-errhandler&gt;

Note that MPI does not guarantee that an MPI program can continue past an error.

See the MPI man page &lt;#open-mpi&gt; for a full list of MPI error codes &lt;#open-mpi-errors&gt;.

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

Note that per the “Return Status” section in the “Point-to-Point Communication” chapter in the MPI Standard &lt;https://www.mpi-forum.org/docs/&gt;, MPI errors on messages queried by MPI\_Probe do not set the **status.MPI\_ERROR** field in the returned *status*. The error code is always passed to the back-end error handler and may be passed back to the caller through the return value of MPI\_Probe if the back-end error handler returns it. The pre-defined MPI error handler **MPI\_ERRORS\_RETURN** exhibits this behavior, for example.

**See also:**

- MPI\_Iprobe &lt;#mpi-iprobe&gt;
- MPI\_Cancel &lt;#mpi-cancel&gt;

# [Copyright](#Copyright)

2003-2026, The Open MPI Community

October 11, 2026

---

Document: MPI_Probe(3) from openmpi-cuda-13.4 5.0.11-r1 (Chainguard OS)
