SHMEM_INFO_GET_NAME(3)

✨ NAME ✨

shmem_info_get_name β€” bestie, this routine literally just hands you the vendor defined character string πŸ’– no cap, that's the whole vibe.


🌸 SYNOPSIS 🌸

C or C++:

#include <mpp/shmem.h>

void shmem_info_get_name(char *name)

Fortran:

include 'mpp/shmem.fh'
SHMEM_INFO_GET_NAME(NAME)
CHARACTER *(*)NAME

πŸ’… DESCRIPTION πŸ’…

okay so shmem_info_get_name() πŸ¦‹ β€” this routine returns the vendor defined character string, and the size of that string is set by the constant SHMEM_MAX_NAME_LEN.

here's the deal, and listen closely:

  • the program calling this function is the one who prepares the memory of size SHMEM_MAX_NAME_LEN
  • then the implementation copies the string of size at most SHMEM_MAX_NAME_LEN into it

now the C vs Fortran tea β˜•:

  • in C, the string is terminated by a null character.
  • in Fortran, a string of size less than SHMEM_MAX_NAME_LEN gets padded with blank characters up to size SHMEM_MAX_NAME_LEN.
πŸ’€ and if the implementation copies a string of size greater than SHMEM_MAX_NAME_LEN? that results in an undefined behavior. not cute. not iconic. just chaos. avoid.

also for the record: multiple invocations of the routine in an OpenSHMEM program always return the same string 🫢 consistency queen. and for a given library implementation, the major and minor version returned by these calls is consistent with the compile-time constants defined in its shmem.h.


πŸ‘€ SEE ALSO πŸ‘€


2003-2026, The Open MPI Community


and that's the whole page, bestie 🌈 thanks for reading, mwah πŸ’œ β€” this was the Open MPI version, dated Sep 10, 2026, and the manual page is SHMEM_INFO_GET_NAME(3) 🧚 ✨

πŸ“¦ where this page comes from