17.2.286. MPI_Publish_name

MPI_Publish_name — Publishes a service name associated with a port

17.2.286.1. SYNTAX

17.2.286.1.1. C Syntax

#include <mpi.h>

int MPI_Publish_name(const char *service_name, MPI_Info info,
     const char *port_name)

17.2.286.1.2. Fortran Syntax

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_PUBLISH_NAME(SERVICE_NAME, INFO, PORT_NAME, IERROR)
     CHARACTER*(*)   SERVICE_NAME, PORT_NAME
     INTEGER         INFO, IERROR

17.2.286.1.3. Fortran 2008 Syntax

USE mpi_f08
MPI_Publish_name(service_name, info, port_name, ierror)
     TYPE(MPI_Info), INTENT(IN) :: info
     CHARACTER(LEN=*), INTENT(IN) :: service_name, port_name
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

17.2.286.2. INPUT PARAMETERS

service_name: A service name (string).
info: Options to the name service functions (handle).
port_name: A port name (string).

17.2.286.3. OUTPUT PARAMETER

ierror: Fortran only: Error status (integer).

17.2.286.4. DESCRIPTION

This routine publishes the pair (service_name, port_name) so that an application may retrieve port_name by calling MPI_Lookup_name with service_name as an argument. It is an error to publish the same service_name twice, or to use a port_name argument that was not previously opened by the calling process via a call to MPI_Open_port.

17.2.286.5. INFO ARGUMENTS

The following keys for info are recognized:

Key                   Type      Description
---                   ----      -----------

range                 char *    Scope in which to publish the service
                                name.  See the NAME SCOPE section
                                below.

persistence           char *    How long the published service name is
                                retained.  See below.

The range info key accepts one of two string values:

nspace: Restrict the operation to processes in the same MPI job (PMIx: namespace) as the calling process.

session: Apply the operation across all processes in the same session.

If the info argument is MPI_INFO_NULL, or is a valid info object that does not contain a range key, session scope is used. Because MPI_Publish_name, MPI_Lookup_name, and MPI_Unpublish_name all share this default, a name published with MPI_INFO_NULL is found by a lookup with MPI_INFO_NULL.

Any other value for range results in an error.

The persistence info key accepts one of four string values:

indef: Retain the service name until it is explicitly unpublished.

proc: Retain the service name until the publishing process terminates.

app: Retain the service name until the publishing application: terminates.
session: Retain the service name until the session terminates. This is: the default if the persistence key is not given.

Any other value for persistence results in an error.

17.2.286.6. NAME SCOPE

Open MPI supports two name scopes, selected by the range info key: nspace and session.

nspace scope restricts the (service_name, port_name) pair to processes in the publisher’s own MPI job, i.e., processes sharing the publisher’s PMIx namespace.

session scope makes the pair visible to every process in the same PMIx session. This includes jobs started by separate mpirun invocations that share a persistent DVM or scheduler allocation, as well as jobs created via MPI_Comm_spawn. session is the default scope.

The same scope must be used to publish, look up, and unpublish a given service name. MPI_Unpublish_name returns an error if the service name is not found in the indicated scope.

As an example of the impact of these scoping rules, consider a job started with mpirun – call it “job1”. A process in job1 publishes a service/port pair using nspace scope. A process in a job that job1 subsequently starts via MPI_Comm_spawn – call it “job2” – is in a different namespace, and so cannot retrieve that pair. Had job1 published with session scope (the default), processes in job2 could retrieve it, as could processes in any other job sharing the same session.

17.2.286.7. 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 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 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 Returns an error code to the application.

MPI applications can also implement their own error handlers by calling:

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

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.