SHMEM_PUT_NBI(3)

💖 SHMEM_PUT_NBI(3) 💖

okay so bestie, we're in the Open MPI manual and this page is all about the nonblocking put routines — shmem_putmem_nbi(3), shmem_char_put_nbi(3), shmem_short_put_nbi(3), shmem_int_put_nbi(3), shmem_long_put_nbi(3), shmem_longlong_put_nbi(3), shmem_float_put_nbi(3), shmem_double_put_nbi(3), shmem_longdouble_put_nbi(3), shmem_put8_nbi(3), shmem_put16_nbi(3), shmem_put32_nbi(3), shmem_put64_nbi(3), shmem_put128_nbi(3). and what do they even do?? they give you a way to copy data from a contiguous local data object over to a data object on a specified PE. that's the whole vibe. ✨

---

🦋 SYNOPSIS 🦋

C or C++

#include <mpp/shmem.h>

void shmem_putmem_nbi(void *dest, const void *source,
  size_t nelems, int pe);

void shmem_char_put(char *dest, const char *source,
  size_t nelems, int pe);

void shmem_short_put(short *dest, const short *source,
  size_t nelems, int pe);

void shmem_int_put(int *dest, const int *source,
  size_t nelems, int pe);

void shmem_long_put(long *dest, const long *source,
  size_t nelems, int pe);

void shmem_longlong_put(long long *dest, const long long *source,
  size_t nelems, int pe);

void shmem_float_put(float *dest, const float *source,
  size_t nelems, int pe);

void shmem_double_put(double *dest, const double *source,
  size_t nelems, int pe);

void shmem_longdouble_put(long double *dest, const long double *source,
  size_t nelems, int pe);

void shmem_put8(void *dest, const void *source,
  size_t nelems, int pe);

void shmem_put16(void *dest, const void *source,
  size_t nelems, int pe);

void shmem_put32(void *dest, const void *source,
  size_t nelems, int pe);

void shmem_put64(void *dest, const void *source,
  size_t nelems, int pe);

void shmem_put128(void *dest, const void *source,
  size_t nelems, int pe);

👀 not me pointing at the synopsis — write these exactly as they appear, besties. the C prototypes above are literally printed that way on the page, so copy them character for character. no "fixing" the names, no adding _nbi where it isn't, got it? iykyk 💅

Fortran

INCLUDE "mpp/shmem.fh"

INTEGER nelems, pe

CALL SHMEM_PUTMEM_NBI(dest, source, nelems, pe)

CALL SHMEM_CHARACTER_PUT_NBI(dest, source, nelems, pe)

CALL SHMEM_COMPLEX_PUT_NBI(dest, source, nelems, pe)

CALL SHMEM_DOUBLE_PUT_NBI(dest, source, nelems, pe)

CALL SHMEM_INTEGER_PUT_NBI(dest, source, nelems, pe)

CALL SHMEM_LOGICAL_PUT_NBI(dest, source, nelems, pe)

CALL SHMEM_REAL_PUT_NBI(dest, source, nelems, pe)

CALL SHMEM_PUT4_NBI(dest, source, nelems, pe)

CALL SHMEM_PUT8_NBI(dest, source, nelems, pe)

CALL SHMEM_PUT32_NBI(dest, source, nelems, pe)

CALL SHMEM_PUT64_NBI(dest, source, nelems, pe)

CALL SHMEM_PUT128_NBI(dest, source, nelems, pe)

---

🌸 DESCRIPTION 🌸

okay so here's the tea 🍓 — these routines return right after posting the operation. like, they hand it off and dip. the operation is considered complete only after a subsequent call to shmem_quiet. at the completion of shmem_quiet, the data has been copied into the dest array on the destination PE. 👑

and now the part you cannot skip: the delivery of data words into the data object on the destination PE may occur in any order. and on top of that, two successive put routines may deliver data out of order unless a call to shmem_fence is introduced between the two calls. so if order matters to you, bestie, use the fence. no cap. 🔥

the arguments are as follows:

• dest — the data object to be updated on the remote PE. this data object must be remotely accessible. non-negotiable, main character requirement.
• source — the data object containing the data to be copied. the source of the slay, if you will.
• nelems — number of elements in the dest and source arrays. nelems must be of type size_t for C. if you're using Fortran, it must be a constant, variable, or array element of default integer type. 📏
• pe — PE number of the remote PE. pe must be of type integer. if you're using Fortran, it must be a constant, variable, or array element of default integer type. 🎯

if you're using Fortran, data types must be of default size. for example, a real variable must be declared as REAL, REAL*4, or REAL(KIND=4). that's just the rule bestie, don't fight it 💜

---

🧚 NOTES 🧚

see intro_shmem(3) for a definition of the term remotely accessible — yeah, it's a whole thing, go read it. 🦋

---

⭐ EXAMPLES ⭐

consider this simple example for C. it's giving classic, it's giving textbook, and honestly? it ate.

#include <stdio.h>
#include <mpp/shmem.h>

main()
{
  long source[10] = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
  static long target[10];
  shmem_init();

  if (shmem_my_pe() == 0) {
    /* put 10 words into target on PE 1 */
    shmem_long_put_nbi(target, source, 10, 1);
    shmem_quiet();
  }
  shmem_barrier_all();  /* sync sender and receiver */
  if (shmem_my_pe() == 1)
    shmem_udcflush();  /* not required on Altix systems */
  printf("target[0] on PE %d is %d\n", shmem_my_pe(), target[0]);
}

---

🌈 SEE ALSO 🌈

• intro_shmem(3)
• shmem_quiet(3)

---

and that's the page, bestie 🫶 this iconic little writeup came from gitclone, marked as an Unreleased developer copy — so yeah, developer-copy energy only, handle with care and go forth and put nonblocking. periodt. 🎀💖