SHMEM_LONGLONG_MAX_TO_ALL(3)

📦

version 5.0.10 · 4 versions

🌸 shmem_longlong_max_to_all — like, the whole `SHMEM_LONGLONG_MAX_TO_ALL(3)` situation 💖

okay bestie, buckle up. this is the Open MPI manual for shmem_longlong_max_to_all and its whole little friend group of max-reduction routines 🫶 same page, same rules, eleven-ish cousins:

shmem_double_max_to_all(3)
shmem_float_max_to_all(3)
shmem_int_max_to_all(3)
shmem_int4_max_to_all(3)
shmem_int8_max_to_all(3)
shmem_long_max_to_all(3)
shmem_longdouble_max_to_all(3)
shmem_longlong_max_to_all(3)
shmem_real4_max_to_all(3)
shmem_real8_max_to_all(3)
shmem_real16_max_to_all(3)
shmem_short_max_to_all(3)

and what do they all do?? they perform a maximum function reduction across a set of processing elements (PEs) ✨ like, they all get together, compare notes, and the biggest value wins. main character energy fr fr 👑

✨ SYNOPSIS ✨

C or C++:

#include <mpp/shmem.h>

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

Fortran besties:

INCLUDE "mpp/shmem.fh"

INTEGER pSync(SHMEM_REDUCE_SYNC_SIZE)

INTEGER nreduce, PE_start, logPE_stride, PE_size

CALL SHMEM_INT4_MAX_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_INT8_MAX_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL4_MAX_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL8_MAX_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL16_MAX_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 is basically an associative binary operation performed across a set of values. wanna see the rest of the SHMEM reduction fam? peek at intro_shmem(3) 🌈

and like every SHMEM collective routine ever, each of these assumes that only the PEs in the active set call it. if a PE that is not in the active set calls a SHMEM collective routine — > undefined behavior results. no cap, don't do that 💀

the nreduce argument is the number of separate reductions you wanna perform. the source array on all PEs in the active set hands over one element per reduction. then 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, iconically 💅

the source and target arrays may be the same array — but they may not be overlapping arrays. periodt.

now the arguments, one by one:

target — a symmetric array of length nreduce elements, there to receive the results of the reduction operations 🎀 the data type of target varies with which version of the reduction routine you're calling. from C? go look at the SYNOPSIS section for the data type tea.

from Fortran, the target data types go like this:

shmem_comp8_max_to_all: Complex, with an element size equal to two 8-byte real values.
shmem_int4_max_to_all: Integer, with an element size of 4 bytes.
shmem_int8_max_to_all: Integer, with an element size of 8 bytes.
shmem_real4_max_to_all: Real, with an element size of 4 bytes.
shmem_real16_max_to_all: Real, with an element size of 16 bytes.
source — a symmetric array of length nreduce elements holding one element for each separate reduction operation. source must have the same data type as target. no mixing, bestie 🙈
nreduce — the number of elements in the target and source arrays. nreduce must be of type integer. if you're doing 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. pWrk must have the same data type as target. in C/C++, this contains max(nreduce/2 + 1, _SHMEM_REDUCE_MIN_WRKDATA_SIZE) elements. in Fortran, this contains 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. if you're using Fortran, it must be a default integer value. every single element of this array has to be initialized with the value _SHMEM_SYNC_VALUE (in C/C++) or SHMEM_SYNC_VALUE (in Fortran) before any of the PEs in the active set enter the reduction routine. do not skip this step asdfghjkl 😭

the values of the arguments 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 all PEs in the active set. everyone matching, everyone coordinated 💞

before any PE calls a reduction routine, you've gotta make sure these conditions exist (synchronization via a barrier or some other method is often needed to keep it that way, jsyk):

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.

and once a reduction routine returns, here's what's 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 in intro_shmem(3). and good news: 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 🎉

you must ensure that 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. watch out for these situations, fr:

if the pSync array is initialized at run time, some type of synchronization is needed to make sure that 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, that can only be assured by doing some type of synchronization. but — special case alert 🚨 — if reduction routines are called with the same active set, you can allocate two pSync and pWrk arrays and alternate between them on successive calls. alternating, it's giving strategic 💅

🌈 EXAMPLES 🌈

Example 1: this Fortran example statically initializes the pSync array and finds the maximum value of real variable FOO across all even PEs.

INCLUDE "mpp/shmem.fh"

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

IF ( MOD(MY_PE(),2) .EQ. 0) THEN
  CALL SHMEM_REAL8_MAX_TO_ALL(FOOMAX, FOO, NR, 0, 1, N$PES/2,
  & PWRK, PSYNC)
  PRINT *, 'Result on PE ', MY_PE(), ' is ', FOOMAX
ENDIF

Example 2: consider the following C/C++ call:

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

the call above is more efficient than — but semantically equivalent to — the combination of the following calls:

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

note that two sets of pWrk and pSync arrays are used alternately, because no synchronization is done between calls 👀 (see? we told you about the alternating thing.)

👑 Copyright 👑

2003-2026, The Open MPI Community

okay that's the whole page diva 💖 you made it to the end, so proud of you. this was SHMEM_LONGLONG_MAX_TO_ALL(3) from the Open MPI crew, last touched Sep 10, 2026 — go forth and reduce, bestie, and remember: initialize your pSync or it's giving undefined behavior 🦋✨

📦 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