SHMEM_LONGLONG_AND_TO_ALL(3)

📦
version 26.9 · 7 versions
, in arch rolling

💅 SHMEM_LONGLONG_AND_TO_ALL(3) — the Open MPI glow-up

okay bestie, so this is a whole family of Open MPI manpages and they all live together, so let me introduce the squad first:

and what do they all do? literally the same thing, just different sizes: performs a bitwise AND operation on symmetric arrays over the active set of PEs. that's it. that's the whole personality. 💖

✨ SYNOPSIS ✨

C or C++

#include <mpp/shmem.h>

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

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

(yes, shmem_long_and_to_all really does have that extra space in int logPE_stride — iconic of it, no notes, we keep it 🫶)

Fortran

INCLUDE "mpp/shmem.fh"

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

CALL SHMEM_INT4_AND_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_INT8_AND_TO_ALL(target, source, nreduce,
& PE_start, logPE_stride, PE_size, pWrk, pSync)

🦋 DESCRIPTION 🦋

so the SHMEM reduction routines compute one or more reductions across symmetric arrays on multiple virtual PEs. a reduction is just… performing an associative binary operation across a set of values. cute, right? if you want the full list of other SHMEM reduction routines, go see intro_shmem(3) — it's the group chat of reduction lore.

the nreduce argument is literally the number of separate reductions you're doing. 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. and the active set? that's the trio PE_start, logPE_stride, PE_size. that's the vip list. 🎀

source and target arrays may be the same array — but they may not be overlapping arrays, and that's not a suggestion, that's a rule. also, like every SHMEM collective routine, each of these assumes only PEs in the active set call the routine. if a PE not in the active set calls a SHMEM collective routine… undefined behavior results. 💀 don't do it.

the arguments, one by one

  • target — a symmetric array, of length nreduce elements, that receives the result of the reduction operations. the data type of target varies with the version of the reduction routine you're calling. calling from C/C++? check the SYNOPSIS for the data type tea ☕. calling from Fortran? the target data types are:
    • shmem_int8_and_to_all: Integer, with an element size of 8 bytes
    • shmem_int4_and_to_all: Integer, with an element size of 4 bytes
  • 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. no mismatching, bestie, it's giving commitment issues.
  • nreduce — the number of elements in the target and source arrays. nreduce must be of type integer. if you're using 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. if you're using 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. if you're using Fortran, it must be a default integer value. (yes it's log base 2, no you may not ask why 🌸)
  • PE_size — the number of PEs in the active set. PE_size must be of type integer. if you're using 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 must be of type long and size _SHMEM_REDUCE_SYNC_SIZE. in Fortran, pSync must be of type integer and size SHMEM_REDUCE_SYNC_SIZE. if you're using Fortran, it must be a default integer value. every element of this array must 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. before!! ⭐

the matching-outfits rule 👑

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, and the same pWrk and pSync work arrays, must be passed to all PEs in the active set. everybody has to be wearing the same fit, fr fr.

before you call — the vibe check

before any PE calls a reduction routine, you must ensure these conditions exist (synchronization via a barrier or some other method is often needed to ensure 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.

after you return — what's true on the local PE

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 in intro_shmem(3). all SHMEM reduction routines reset the values in pSync before they return, so a particular pSync buffer need only be initialized the first time it is used. (one and done, we love a low-maintenance icon ✨)

you must ensure that the pSync array is not being updated on any PE in the active set while any of the PEs participate in processing of a SHMEM reduction routine. be careful to avoid these situations:

  • if the pSync array is initialized at run time, some type of synchronization is needed to ensure 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, this can be assured only by doing some type of synchronization. but here's the cute trick 🌈: in the special case of reduction routines being called with the same active set, you can allocate two pSync and pWrk arrays and alternate between them on successive calls. two outfits, alternating. main character energy.

🌸 EXAMPLES 🌸

Example 1: this Fortran example statically initializes the pSync array and finds the logical AND of the integer 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 PWRK(MAX(NR/2+1, SHMEM_REDUCE_MIN_WRKDATA_SIZE))
INTEGER FOO, FOOAND
COMMON /COM/ FOO, FOOAND, PWRK
INTRINSIC MY_PE

IF ( MOD(MY_PE(),2) .EQ. 0) THEN
  CALL SHMEM_INT8_AND_TO_ALL(FOOAND, FOO, NR, 0, 1, N$PES/2,
  & PWRK, PSYNC)
  PRINT *, 'Result on PE ', MY_PE(), ' is ', FOOAND
ENDIF

Example 2: consider the following C call:

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

the preceding call is more efficient, but semantically equivalent to, the combination of the following calls:

shmem_int_and_to_all(&(target[0]), &(source[0]), 1, 0, 0, 8,
  pwrk1, psync1);

shmem_int_and_to_all(&(target[1]), &(source[1]), 1, 0, 0, 8,
  pwrk2, psync2);

shmem_int_and_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? told you the two-outfit trick was real. 👑

🎀 SEE ALSO:

2003-2026, The Open MPI Community

and that's the whole page, bestie — this one's from the Open MPI crew, dated May 14, 2026, filed under SHMEM_LONGLONG_AND_TO_ALL(3). go reduce something pretty 🦋✨🫶

📦 where this page comes from