--- name: "MPI_Comm_split_type.openmpi" section: "3" platform: "linux" package: "openmpi-doc" version: "5.0.7-1" release: "Debian 13 (Trixie)" platforms: ["linux"] url: "https://new.carta.tech/MPI_Comm_split_type.openmpi" --- # MPI\_Comm\_split\_type.openmpi(3) *MPI\_Comm\_split\_type* — Creates new communicators based on colors and keys. ## SYNTAX ### C Syntax ``` #include int MPI_Comm_split_type(MPI_Comm comm, int split_type, int key, MPI_Info info, MPI_Comm *newcomm) ``` ### Fortran Syntax ``` USE MPI ! or the older form: INCLUDE 'mpif.h' MPI_COMM_SPLIT_TYPE(COMM, SPLIT_TYPE, KEY, INFO, NEWCOMM, IERROR) INTEGER COMM, SPLIT_TYPE, KEY, INFO, NEWCOMM, IERROR ``` ### Fortran 2008 Syntax ``` USE mpi_f08 MPI_Comm_split_type(comm, split_type, key, info, newcomm, ierror) TYPE(MPI_Comm), INTENT(IN) :: comm INTEGER, INTENT(IN) :: split_type, key TYPE(MPI_Info), INTENT(IN) :: info TYPE(MPI_Comm), INTENT(OUT) :: newcomm INTEGER, OPTIONAL, INTENT(OUT) :: ierror ``` ## INPUT PARAMETERS - • **comm**: Communicator (handle). - • **split\_type**: Type of processes to be grouped together (integer). - • **key**: Control of rank assignment (integer). - • **info**: Info argument (handle). ## OUTPUT PARAMETERS - • **newcomm**: New communicator (handle). - • **ierror**: Fortran only: Error status (integer). ## DESCRIPTION This function partitions the group associated with *comm* into disjoint subgroups, based on the type specified by *split\_type*. Each subgroup contains all processes of the same type. Within each subgroup, the processes are ranked in the order defined by the value of the argument *key*, with ties broken according to their rank in the old group. A new communicator is created for each subgroup and returned in newcomm. This is a collective call; all processes must provide the same *split\_type*, but each process is permitted to provide different values for key. An exception to this rule is that a process may supply the type value MPI\_UNDEFINED, in which case newcomm returns MPI\_COMM\_NULL. ## SPLIT TYPES - **MPI\_COMM\_TYPE\_SHARED** This type splits the communicator into subcommunicators, each of which can create a shared memory region. - **OMPI\_COMM\_TYPE\_NODE** Synonym for MPI\_COMM\_TYPE\_SHARED. - **OMPI\_COMM\_TYPE\_HWTHREAD** This type splits the communicator into subcommunicators, each of which belongs to the same hardware thread. - **OMPI\_COMM\_TYPE\_CORE** This type splits the communicator into subcommunicators, each of which belongs to the same core/processing unit. - **OMPI\_COMM\_TYPE\_L1CACHE** This type splits the communicator into subcommunicators, each of which belongs to the same L1 cache. - **OMPI\_COMM\_TYPE\_L2CACHE** This type splits the communicator into subcommunicators, each of which belongs to the same L2 cache. - **OMPI\_COMM\_TYPE\_L3CACHE** This type splits the communicator into subcommunicators, each of which belongs to the same L3 cache. - **OMPI\_COMM\_TYPE\_SOCKET** This type splits the communicator into subcommunicators, each of which belongs to the same socket. - **OMPI\_COMM\_TYPE\_NUMA** This type splits the communicator into subcommunicators, each of which belongs to the same NUMA-node. - **OMPI\_COMM\_TYPE\_BOARD** This type splits the communicator into subcommunicators, each of which belongs to the same board. - **OMPI\_COMM\_TYPE\_HOST** This type splits the communicator into subcommunicators, each of which belongs to the same host. - **OMPI\_COMM\_TYPE\_CU** This type splits the communicator into subcommunicators, each of which belongs to the same computational unit. - **OMPI\_COMM\_TYPE\_CLUSTER** This type splits the communicator into subcommunicators, each of which belongs to the same cluster. ## NOTES The communicator keys denoted with an *OMPI\_* prefix instead of an *MPI\_* prefix are specific to Open MPI, and are not part of the MPI standard. Their use should be protected by the *OPEN\_MPI* C preprocessor macro. ## 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: - • *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. 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. **SEE ALSO:** - • *MPI\_Comm\_create* - • *MPI\_Intercomm\_create* - • *MPI\_Comm\_dup* - • *MPI\_Comm\_free* - • *MPI\_Comm\_split* ## COPYRIGHT 2003-2025, The Open MPI Community