SHMEM_COLLECT(3)SHMEM_COLLECT(3)NAME
shmem_collect4, shmem_collect8, shmem_collect32, shmem_collect64,
shmem_fcollect, shmem_fcollect4, shmem_fcollect8, shmem_fcollect32,
shmem_fcollect64 - Concatenates blocks of data from multiple
processing elements (PEs) to an array in every PE
SYNOPSIS
C or C++:
#include <mpp/shmem.h>
void shmem_collect32(void *target, const void *source, size_t
nlong, int PE_start, int logPE_stride, int PE_size, long *pSync);
void shmem_collect64(void *target, const void *source, size_t
nlong, int PE_start, int logPE_stride, int PE_size, long *pSync);
void shmem_fcollect32(void *target, const void *source, size_t
nlong, int PE_start, int logPE_stride, int PE_size, long *pSync);
void shmem_fcollect64(void *target, const void *source, size_t
nlong, int PE_start, int logPE_stride, int PE_size, long *pSync);
Fortran:
INCLUDE "mpp/shmem.fh"
INTEGER nlong
INTEGER PE_start, logPE_stride, PE_size
INTEGER pSync(SHMEM_COLLECT_SYNC_SIZE)
CALL SHMEM_COLLECT4(target, source, nlong, PE_start, logPE_stride,
PE_size, pSync)
CALL SHMEM_COLLECT8(target, source, nlong, PE_start, logPE_stride,
PE_size, pSync)
CALL SHMEM_FCOLLECT4(target, source, nlong, PE_start, logPE_stride,
PE_size, pSync)
CALL SHMEM_FCOLLECT8(target, source, nlong, PE_start, logPE_stride,
PE_size, pSync)
DESCRIPTION
The shared memory (SHMEM) collect and fcollect routines concatenate
nlong 64- or 32-bit data items from the source array into the target
array, over the set of PEs defined by PE_start, log2PE_stride, and
PE_size, in processor number order. The resultant target array
contains the contribution from PE PE_start first, then the
contribution from PE PE_start + PE_stride second, and so on. The
collected result is written to the target array for all PEs in the
active set.
The fcollect routines require that nlong be the same value in all
participating PEs, while the collect routines allow nlong to vary from
PE to PE.
The resulting target array is as follows:
----------------------------------------------------------
source(1..nlong)
from PE (PE_start + 0 * (2**logPE_stride))
----------------------------------------------------------
source(1..nlong)
from PE (PE_start + 1 * (2**logPE_stride))
----------------------------------------------------------
...
----------------------------------------------------------
source(1..nlong) from
PE (PE_start + (PE_size - 1) * (2**logPE_stride))
----------------------------------------------------------
As with all SHMEM collective routines, each of these routines assumes
that 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.
The arguments are as follows:
target A symmetric array. The target argument must be large
enough to accept the concatenation of the source arrays
on all PEs. The data types are as follows:
* For shmem_collect8, shmem_collect64, shmem_fcollect8,
and shmem_fcollect64, any data type with an element
size of 64 bits. Fortran derived types, Fortran
character type, and C/C++ structures are not permitted.
* For shmem_collect4, shmem_collect32, shmem_fcollect4,
and shmem_fcollect32, any data type with an element
size of 32 bits. Fortran derived types, Fortran
character type, and C/C++ structures are not permitted.
source A symmetric data object that can be of any type
permissible for the target argument.
nlong The number of elements in the source array. nlong 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.
pSync A symmetric work array.
In C/C++, pSync must be of type int and size
_SHMEM_COLLECT_SYNC_SIZE.
In Fortran, pSync must be of type integer and size
SHMEM_COLLECT_SYNC_SIZE. If you are 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
shmem_barrier().
The values of arguments 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 pSync work array must be passed to all PEs in the active
set.
Upon return from a collective routine, the following 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 and symmetric are defined in intro_shmem(3).
All SHMEM collective routines reset the values in pSync before they
return, so a particular pSync buffer need only be initialized the
first time it is used.
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 collective 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 array can be reused on a subsequent SHMEM collective routine
only if none of the PEs in the active set are still processing a
prior SHMEM collective routine call that used the same pSync array.
In general, this may be ensured only by doing some type of
synchronization. However, in the special case of SHMEM routines
being called with the same active set, you can allocate two pSync
arrays and alternate between them on successive calls.
The collective routines operate on active PE sets that have a
non-power-of-two PE_size with some performance degradation. They
operate with no performance degradation when nlong is a
non-power-of-two value.
EXAMPLES
C/C++:
for (i=0; i < _SHMEM_COLLECT_SYNC_SIZE; i++) {
pSync[i] = _SHMEM_SYNC_VALUE;
}
shmem_barrier_all(); /* Wait for all PEs to initialize pSync */
shmem_collect32(target, source, 64, pe_start, logPE_stride,
pe_size, pSync);
Fortran:
INTEGER PSYNC(SHMEM_COLLECT_SYNC_SIZE)
DATA PSYNC /SHMEM_COLLECT_SYNC_SIZE*SHMEM_SYNC_VALUE/
CALL SHMEM_COLLECT4(TARGET, SOURCE, 64, PE_START, LOGPE_STRIDE,
& PE_SIZE, PSYNC)
SEE ALSOintro_shmem(3)
