omp_nested(3)

NAME
     omp_set_nested, omp_get_nested, OMP_SET_NESTED, OMP_GET_NESTED -
     Manipulates or reports status of nested parallelism

SYNOPSIS
     C/C++:
          #integer <omp_set_nested>
          void omp_set_nested(integer-expression)

     Fortran:
          CALL OMP_SET_NESTED(logical-expression)

IMPLEMENTATION
     UNICOS and IRIX systems

STANDARDS
     OpenMP Fortran API
     OpenMP C/C++ API

DESCRIPTION
     Fortran differences are noted in parentheses in the following
     descriptions.

     omp_set_nested, OMP_SET_NESTED
          This procedure enables or disables nested parallelism.

     If integer-expression (logical-expression) evaluates to zero (FALSE),
     which is the default, nexted parallelism is disabled and nested
     parallel regions are seialized and executed by the current thread.  If
     set to non-zero (TRUE), nested parallelism is enabled and parallel
     regions that are nested may deploy additional threads to form the
     team.  This call has precedence over the OMP_NESTED environment
     variable.  When nested parallelism is enabled, the number of threads
     used to execute the nested parallel regions are implementation
     dependent.  As a result, implementations that are compliant to OpenMP
     are allowed to serialize nested parallel regions even when nested
     parallelism is enabled.

ENVIRONMENT VARIABLES
     OMP_NESTED
          This variable enables or disables nested parallelism.  If set to
          TRUE, nested parallelism is enabled; if it is set to FALSE, it is
          disabled.  The default value is FALSE.

NOTES
     For the OpenMP directives to be honored by the compiler, you must
     specify -mp on your compiler command line explicitly.  In addition,
     -MP:open_mp=ON, which is the default setting, must be in effect.

SEE ALSO
     omp_lock(3) on how to manipulate locks

     omp_threads(3) for runtime library procedures used to set, call or
     return numbers of threads

     This man page is available only online.