SHMEM_ADD(3) | Open MPI | SHMEM_ADD(3) |
shmem_int_add(3), shmem_int4_add(3), shmem_int8_add(3), shmem_long_add(3), shmem_longlong_add(3) - Performs an atomic add operation.
C or C++:
#include <mpp/shmem.h> void shmem_int_add(int *target, int value, int pe); void shmem_long_add(long *target, long value, int pe); void shmem_longlong_add(long long *target, long long value,
int pe);
Fortran:
include 'mpp/shmem.h' INTEGER pe CALL SHMEM_INT4_ADD(target, value, pe) CALL SHMEM_INT8_ADD(target, value, pe)
The atomic add routines add value to the data at address target on PE pe. The operation completes without the possibility of another process updating target between the time of the fetch and the update.
The arguments are as follows:
The term remotely accessible is defined in intro_shmem(3).
October 29, 2018 | 3.1.3 |