SHMEM_LONGDOUBLE_SUM_TO_ALL(3)

📦

version 5.0.10 · 4 versions

💖 SHMEM_LONGDOUBLE_SUM_TO_ALL 💖 — Open MPI, section 3

okay bestie, so this page is a whole family of functions and they all do the same cute thing: a sum reduction across a set of processing elements (PEs). This is part 1 of 2, so buckle up 🎀

✨ NAME ✨

The icons in question, all of them, no cap:

shmem_comp4_sum_to_all
shmem_comp8_sum_to_all
shmem_complexd_sum_to_all
shmem_complexf_sum_to_all
shmem_double_sum_to_all
shmem_float_sum_to_all
shmem_int_sum_to_all
shmem_int4_sum_to_all
shmem_int8_sum_to_all
shmem_long_sum_to_all
shmem_longdouble_sum_to_all
shmem_longlong_sum_to_all
shmem_real4_sum_to_all
shmem_real8_sum_to_all
shmem_real16_sum_to_all
shmem_short_sum_to_all

…and what do they do? Performs a sum reduction across a set of processing elements (PEs). that's the whole vibe 💅

🦋 SYNOPSIS 🦋

C or C++

#include <mpp/shmem.h>

void shmem_complexd_sum_to_all(double complex *target,
  const double complex *source, int nreduce, int PE_start,
  int logPE_stride, int PE_size, double complex *pWrk,
  long *pSync)

void shmem_complexf_sum_to_all(float complex *target,
  const float complex *source, int nreduce, int PE_start,
  int logPE_stride, int PE_size, float complex *pWrk,
  long *pSync)

void shmem_double_sum_to_all(double *target,
  const double *source, int nreduce, int PE_start, int logPE_stride,
  int PE_size, double *pWrk, long *pSync)

void shmem_float_sum_to_all(float *target, const float *source,
  int nreduce, int PE_start, int logPE_stride, int PE_size,
  float *pWrk, long *pSync)

void shmem_int_sum_to_all(int *target, const int *source,
  int nreduce, int PE_start, int logPE_stride, int PE_size,
  int *pWrk, long *pSync)

void shmem_long_sum_to_all(long *target, const long *source,
  int nreduce, int PE_start, int logPE_stride, int PE_size,
  long *pWrk, long *pSync)

void shmem_longdouble_sum_to_all(long double *target,
  const long double *source, int nreduce, int PE_start, int
  logPE_stride, int PE_size, long double *pWrk, long *pSync)

void shmem_longlong_sum_to_all(long long *target,
  const long long *source, int nreduce, int PE_start,
  int logPE_stride, int PE_size, long long *pWrk,
  long *pSync)

void shmem_short_sum_to_all(short *target, const short *source,
  int nreduce, int PE_start, int logPE_stride, int PE_size,
  short *pWrk, long *pSync)

Fortran

INCLUDE "mpp/shmem.fh"

INTEGER pSync(SHMEM_REDUCE_SYNC_SIZE)
INTEGER nreduce, PE_start, logPE_stride, PE_size

CALL SHMEM_COMP4_SUM_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_COMP8_SUM_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_INT4_SUM_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_INT8_SUM_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL4_SUM_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL8_SUM_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL16_SUM_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

🌸 DESCRIPTION 🌸

So. The shared memory (SHMEM) reduction routines compute one or more reductions across symmetric arrays on multiple virtual PEs. A reduction performs an associative binary operation across a set of values. Wanna see the other SHMEM reduction routines? They're all listed in intro_shmem(3), go look 👀

Same as literally every SHMEM collective routine: each of these assumes that only PEs in the active set call the routine. If a PE that is not in the active set calls a SHMEM collective routine, you get undefined behavior. periodt. don't do it 🙈

The nreduce argument decides how many separate reductions happen. The source array on all PEs in the active set hands over one element for each reduction. The results land in the target array on all PEs in the active set. The active set itself is defined by the PE_start, logPE_stride, PE_size triplet. slay.

The source and target arrays may be the same array, but they may not be overlapping arrays. okay? okay 💅

The arguments are as follows:

target — The remotely accessible integer data object to be updated on the remote PE. In C/C++, the type of target should match whatever the SYNOPSIS section implies. If you're on the Fortran compiler, it must be of type integer with an element size of 4 bytes for SHMEM_INT4_ADD and 8 bytes for SHMEM_INT8_ADD.
value — The value to be atomically added to target. In C/C++, the type of value should match what the SYNOPSIS section implies. In Fortran, it must be of type integer with an element size of target.
pe — An integer that indicates the PE number upon which target is to be updated. In Fortran it must be a default integer value.
target — A symmetric array of length nreduce elements that receives the results of the reduction operations. The data type of target varies with the version of the reduction routine being called and the language used. From C/C++, peep the SYNOPSIS section for data type info. From Fortran, the target data types are:
- shmem_comp4_sum_to_all: COMPLEX(KIND=4).
- shmem_comp8_sum_to_all: Complex. In Fortran it must be a default complex value.
- shmem_int4_sum_to_all: INTEGER(KIND=4).
- shmem_int8_sum_to_all: Integer. In Fortran it must be a default integer value.
- shmem_real4_sum_to_all: REAL(KIND=4).
- shmem_real8_sum_to_all: Real. In Fortran it must be a default real value.
- shmem_real16_sum_to_all: Real. In Fortran it must be a default real value.
source — A symmetric array of length nreduce elements containing one element for each separate reduction operation. source must have the same data type as target.
nreduce — The number of elements in the target and source arrays. nreduce must be of type integer. In Fortran it must be a default integer value.
PE_start — The lowest virtual PE number of the active set of PEs. PE_start must be of type integer. In Fortran it must be a default integer value.
logPE_stride — The log (base 2) of the stride between consecutive virtual PE numbers in the active set. logPE_stride must be of type integer. In Fortran it must be a default integer value.
PE_size — The number of PEs in the active set. PE_size must be of type integer. In Fortran it must be a default integer value.
pWrk — A symmetric work array. It must have the same data type as target. In C/C++ this holds max(nreduce/2 + 1, _SHMEM_REDUCE_MIN_WRKDATA_SIZE) elements. In Fortran it holds max(nreduce/2 + 1, SHMEM_REDUCE_MIN_WRKDATA_SIZE) elements.
pSync — A symmetric work array. In C/C++ pSync is of type long and size _SHMEM_REDUCE_SYNC_SIZE. In Fortran pSync is of type integer and size SHMEM_REDUCE_SYNC_SIZE, and it must be a default integer value. Every single element of this array must be initialized with _SHMEM_SYNC_VALUE (C/C++) or SHMEM_SYNC_VALUE (Fortran) before any PE in the active set enters the reduction routine. no skipping this bestie, it's giving mandatory 💀

The values of nreduce, PE_start, logPE_stride, and PE_size must be equal on all PEs in the active set. And the same target and source arrays, plus the same pWrk and pSync work arrays, must be passed to every PE in the active set. matching fits, matching energy ✨

Before any PE calls a reduction routine, you have to make sure these conditions exist (synchronization via a barrier or some other method is often needed to make sure of this):

The pWrk and pSync arrays on all PEs in the active set are not still in use from a prior call to a collective SHMEM routine.
The target array on all PEs in the active set is ready to accept the results of the reduction.

Upon return from a reduction routine, these are true for the local PE:

The target array is updated. 🎉
The values in the pSync array are restored to the original values.

👑 NOTES 👑

The terms collective, symmetric, and cache aligned are defined over in intro_shmem(3) — go read that one too, it's giving required reading 📚

All SHMEM reduction routines reset the values in pSync before they return, so a particular pSync buffer only needs to be initialized the first time it's used. one and done, iconic.

🚨 You must make sure the pSync array is not being updated on any PE in the active set while any of the PEs are participating in processing of a SHMEM reduction routine. That's the rule. That's the whole rule.

Be careful of these situations, fr fr:

If the pSync array is initialized at run time, some type of synchronization is needed to make sure all PEs in the working set have initialized pSync before any of them enter a SHMEM routine called with the pSync synchronization array.
A pSync or pWrk array can be reused in a subsequent reduction routine call only if none of the PEs in the active set are still processing a prior reduction routine call that used the same pSync or pWrk arrays.

In general, you can only be sure of that by doing some type of synchronization. However — special case alert 🎀 — when reduction routines are being called with the same active set, you can allocate two pSync and pWrk arrays and alternate between them on successive calls. double-buffering but make it cute 🔥

🌈 EXAMPLES 🌈

Ok bestie, real talk, this is where it all clicks, so stick with me 💖

✨ Example 1 ✨

This Fortran example statically initializes the pSync array, and then it finds the sum of the real variable FOO across all even PEs. Cute, right? Not me being obsessed with a statically initialized sync array 😭

      INCLUDE "mpp/shmem.fh"

      INTEGER PSYNC(SHMEM_REDUCE_SYNC_SIZE)
      DATA PSYNC /SHMEM_REDUCE_SYNC_SIZE*SHMEM_SYNC_VALUE/
      PARAMETER (NR=1)
      REAL FOO, FOOSUM, PWRK(MAX(NR/2+1,SHMEM_REDUCE_MIN_WRKDATA_SIZE))
      COMMON /COM/ FOO, FOOSUM, PWRK
      INTRINSIC MY_PE

      IF ( MOD(MY_PE(),2) .EQ. 0) THEN
        CALL SHMEM_INT4_SUM_TO_ALL(FOOSUM, FOO, NR, 0, 1, N$PES/2,
        & PWRK, PSYNC)
        PRINT *, 'Result on PE ', MY_PE(), ' is ', FOOSUM
      ENDIF

See how the PSYNC array gets set up with SHMEM_REDUCE_SYNC_SIZE and SHMEM_SYNC_VALUE? That's the static init moment, and it ate 💅 Also note the MOD(MY_PE(),2) .EQ. 0 gate — only the even PEs go in here, periodt.

🦋 Example 2 🦋

Now picture this C/C++ call, bestie:

shmem_int_sum_to_all( target, source, 3, 0, 0, 8, pwrk, psync );

This one call is more efficient, but it's semantically equivalent to gluing together these three separate calls down below — like, same meaning, less work, main character energy fr fr:

shmem_int_sum_to_all(&(target[0]), &(source[0]), 1, 0, 0, 8,
  pwrk1, psync1);
shmem_int_sum_to_all(&(target[1]), &(source[1]), 1, 0, 0, 8,
  pwrk2, psync2);
shmem_int_sum_to_all(&(target[2]), &(source[2]), 1, 0, 0, 8,
  pwrk1, psync1);

👀 bestie, important: two sets of pWrk and pSync arrays are used alternately, and the reason is that no synchronization is done between calls. So pwrk1/psync1, then pwrk2/psync2, then pwrk1/psync1 again. Do not skip that detail, it's literally the whole point 😤

See also:

intro_shmem(3)

👑 Copyright 👑

2003-2026, The Open MPI Community

And that's a wrap on SHMEM_LONGDOUBLE_SUM_TO_ALL(3), bestie 🫶 this page came from the Open MPI crew, dated Sep 10, 2026, and yes the title at the bottom of the page is literally the function name in all caps like it's walking the runway 💖 go forth and sum, slay, no cap ✨🎀

📦 where this page comes from

ubuntu stonking: openmpi-doc 5.0.10-6 · other
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash on 2026-11-03: a language model's rewrite of the original, under its terms