1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
|
/*
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2017 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* Copyright (c) 2012 Oak Ridge National Labs. All rights reserved.
* Copyright (c) 2015 Cisco Systems, Inc. All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#include "ompi_config.h"
#include "coll_basic.h"
#include "mpi.h"
#include "ompi/constants.h"
#include "opal/util/bit_ops.h"
#include "ompi/mca/pml/pml.h"
#include "ompi/mca/coll/coll.h"
#include "ompi/mca/coll/base/coll_tags.h"
#include "coll_basic.h"
/*
* barrier_intra_log
*
* Function: - barrier using O(log(N)) algorithm
* Accepts: - same as MPI_Barrier()
* Returns: - MPI_SUCCESS or error code
*/
int
mca_coll_basic_barrier_intra_log(struct ompi_communicator_t *comm,
mca_coll_base_module_t *module)
{
int i;
int err;
int peer;
int dim;
int hibit;
int mask;
int size = ompi_comm_size(comm);
int rank = ompi_comm_rank(comm);
/* Send null-messages up and down the tree. Synchronization at the
* root (rank 0). */
dim = comm->c_cube_dim;
hibit = opal_hibit(rank, dim);
--dim;
/* Receive from children. */
for (i = dim, mask = 1 << i; i > hibit; --i, mask >>= 1) {
peer = rank | mask;
if (peer < size) {
err = MCA_PML_CALL(recv(NULL, 0, MPI_BYTE, peer,
MCA_COLL_BASE_TAG_BARRIER,
comm, MPI_STATUS_IGNORE));
if (MPI_SUCCESS != err) {
return err;
}
}
}
/* Send to and receive from parent. */
if (rank > 0) {
peer = rank & ~(1 << hibit);
err =
MCA_PML_CALL(send
(NULL, 0, MPI_BYTE, peer,
MCA_COLL_BASE_TAG_BARRIER,
MCA_PML_BASE_SEND_STANDARD, comm));
if (MPI_SUCCESS != err) {
return err;
}
err = MCA_PML_CALL(recv(NULL, 0, MPI_BYTE, peer,
MCA_COLL_BASE_TAG_BARRIER,
comm, MPI_STATUS_IGNORE));
if (MPI_SUCCESS != err) {
return err;
}
}
/* Send to children. */
for (i = hibit + 1, mask = 1 << i; i <= dim; ++i, mask <<= 1) {
peer = rank | mask;
if (peer < size) {
err = MCA_PML_CALL(send(NULL, 0, MPI_BYTE, peer,
MCA_COLL_BASE_TAG_BARRIER,
MCA_PML_BASE_SEND_STANDARD, comm));
if (MPI_SUCCESS != err) {
return err;
}
}
}
/* All done */
return MPI_SUCCESS;
}
/*
* barrier_inter_lin
*
* Function: - barrier using O(log(N)) algorithm
* Accepts: - same as MPI_Barrier()
* Returns: - MPI_SUCCESS or error code
*/
int
mca_coll_basic_barrier_inter_lin(struct ompi_communicator_t *comm,
mca_coll_base_module_t *module)
{
int rank;
int result;
rank = ompi_comm_rank(comm);
return comm->c_coll->coll_allreduce(&rank, &result, 1, MPI_INT, MPI_MAX,
comm, comm->c_coll->coll_allreduce_module);
}
|
