SHMEM_INT_PROD_TO_ALL(3)

📦
version 5.0.10 · 4 versions
, in ubuntu stonking

💖 SHMEM_INT_PROD_TO_ALL 💖

okay bestie, gather round 🫶 this is the Open MPI manual page for SHMEM_INT_PROD_TO_ALL(3) — yes the name is a whole mouthful, we're gonna call it "the product-reduction-to-all crew" and honestly? iconic. What it does, in one breath: performs a product reduction across a set of processing elements (PEs). And it's not just one function, it's a whole squad of them:

✨ SYNOPSIS ✨

C or C++:

#include <mpp/shmem.h>

void shmem_complexd_prod_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_prod_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_prod_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_prod_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_prod_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_prod_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_prod_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_prod_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_prod_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_PROD_TO_ALL(target, source, nreduce, PE_start,
& logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_COMP8_PROD_TO_ALL(target, source, nreduce, PE_start,
& logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_INT4_PROD_TO_ALL(target, source, nreduce, PE_start,
& logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_INT8_PROD_TO_ALL(target, source, nreduce, PE_start,
& logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL4_PROD_TO_ALL(target, source, nreduce, PE_start,
& logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL8_PROD_TO_ALL(target, source, nreduce, PE_start,
& logPE_stride, PE_size, pWrk, pSync)

CALL SHMEM_REAL16_PROD_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 does an associative binary operation across a set of values (here: multiplication, go figure 💅). Wanna see the other SHMEM reduction routines? Check intro_shmem(3), they're all listed there.

Like every SHMEM collective routine, each of these assumes only the PEs in the active set call it. And bestie, listen: if a PE that is not in the active set calls a SHMEM collective routine, you get undefined behavior. That's the bad kind of plot twist. 💀

The nreduce argument sets how many separate reductions you're doing. The source array on every PE in the active set hands over one element per reduction, and the results land in the target array on every PE in the active set. The active set is defined by the PE_start, logPE_stride, PE_size triplet — the holy trinity, iykyk 👑

source and target can be the same array, but they may not be overlapping arrays. No cap, that's a hard rule.

The arguments are as follows:

  • target — a symmetric array of length nreduce elements to receive 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. When calling from C/C++, refer to the SYNOPSIS section for data type information. When calling from Fortran, the target data types are as follows:
    • shmem_comp4_prod_to_all: Complex, with an element size equal to two 4-byte real values.
    • shmem_comp8_prod_to_all: Complex, with an element size equal to two 8-byte real values.
    • shmem_int4_prod_to_all: Integer, with an element size of 4 bytes
    • shmem_int8_prod_to_all: Integer, with an element size of 8 bytes
    • shmem_real4_prod_to_all: Real, with an element size of 4 bytes
    • shmem_real8_prod_to_all: Real, with an element size of 8 bytes
    • shmem_real16_prod_to_all: Real, with an element size of 16 bytes
  • source — a symmetric array, of length nreduce elements, that contains one element for each separate reduction operation. source must have the same data type as target — they're matching, periodt.
  • nreduce — the number of elements in the target and source arrays. nreduce must be of type integer. If you are 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 are 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 are using 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. If you are using Fortran, it must be a default integer value.
  • pWrk — a symmetric work array. The pWrk argument 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 are using Fortran, it must be a default integer value. Before any of the PEs in the active set enter the reduction routine, every element of this array must be initialized with the value _SHMEM_SYNC_VALUE (in C/C++) or SHMEM_SYNC_VALUE (in Fortran).

The values of the arguments nreduce, PE_start, logPE_stride, and PE_size must be equal on all PEs in the active set. The same target and source arrays, and the same pWrk and pSync work arrays, must be passed to all PEs in the active set. Before any PE calls a reduction routine, you must ensure that the following 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, and 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, the following are true for the local PE: the target array is updated, and the values in the pSync array are restored to the original values. cute, tidy, respectful 💖

🎀 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, bestie ✨

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 of the following 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. However — special case alert 🚨 — when reduction routines are called with the same active set, you can allocate two pSync and pWrk arrays and alternate between them on successive calls.

fr fr, don't skip the synchronization. This is the part where people's code goes feral and nobody knows why. 🙈

🦋 EXAMPLES 🦋

Example 1: This Fortran example statically initializes the pSync array and finds the product of the real variable FOO across all the 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, FOOPROD, PWRK(MAX(NR/2+1,SHMEM_REDUCE_MIN_WRKDATA_SIZE))
COMMON /COM/ FOO, FOOPROD, PWRK
INTRINSIC MY_PE

IF ( MOD(MY_PE(),2) .EQ. 0) THEN
  CALL SHMEM_COMP8_PROD_TO_ALL(FOOPROD, FOO, NR, 0, 1, N$PES/2,
  & PWRK, PSYNC)
  PRINT *, 'Result on PE ', MY_PE(), ' is ', FOOPROD
ENDIF

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

shmem_short_prod_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_short_prod_to_all(&(target[0]), &(source[0]), 1, 0, 0, 8,
  pwrk1, psync1);
shmem_short_prod_to_all(&(target[1]), &(source[1]), 1, 0, 0, 8,
  pwrk2, psync2);
shmem_short_prod_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? alternating buffers, main character energy 👑

🌈 See also 🌈

2003-2026, The Open MPI Community

and that's the whole page, bestie 🫶 it's giving Open MPI, it's giving Sep 10, 2026, and the copyright is held tight by The Open MPI Community from 2003 all the way through 2026. go multiply some stuff across your PEs and slay 💅✨

📦 where this page comes from