SHMEM\_FINALIZE(3) Open MPI SHMEM\_FINALIZE(3) shmem\_finalize - A collective operation that releases resources used by the OpenSHMEM library. This only terminates the Open-SHMEM portion of a program, not the entire program. # [SYNOPSIS](#SYNOPSIS) C or C++: ``` #include void shmem_finalize(void) ``` Fortran: ``` include 'mpp/shmem.fh' CALL SHMEM_FINALIZE ``` # [DESCRIPTION](#DESCRIPTION) shmem\_finalize is a collective operation that ends the OpenSHMEM portion of a program previously initialized by shmem\_init <#shmem-init> and releases resources used by the OpenSHMEM library. This collective operation requires all PEs to participate in the call. There is an implicit global barrier in shmem\_finalize so that pending communication is completed, and no resources can be released until all PEs have entered shmem\_finalize. shmem\_finalize must be the last OpenSHMEM library call encountered in the OpenSHMEM portion of a program. A call to shmem\_finalize will release any resources initialized by a corresponding call to shmem\_init <#shmem-init>. All processes and threads that represent the PEs will still exist after the call to shmem\_finalize returns, but they will no longer have access to any resources that have been released. **See also:** [*intro\_shmem*(3)](/man/enterprise-packages/openmpi-cuda-13.4/intro_shmem.3) [*shmem\_my\_pe*(3)](/man/enterprise-packages/openmpi-cuda-13.4/shmem_my_pe.3) [*shmem\_init*(3)](/man/enterprise-packages/openmpi-cuda-13.4/shmem_init.3) # [Copyright](#Copyright) 2003-2026, The Open MPI Community September 17, 2026 --- Document: shmem_finalize(3) from openmpi-cuda-13.4 5.0.11-r0 (Chainguard OS)