1 /* Copyright (c) 2013-2019. The SimGrid Team.
2 * All rights reserved. */
4 /* This program is free software; you can redistribute it and/or modify it
5 * under the terms of the license (GNU LGPL) which comes with this package. */
7 #include "../colls_private.hpp"
9 /*****************************************************************************
11 Copyright (c) 2006, Ahmad Faraj & Xin Yuan,
14 Redistribution and use in source and binary forms, with or without
15 modification, are permitted provided that the following conditions are met:
17 * Redistributions of source code must retain the above copyright notice,
18 this list of conditions and the following disclaimer.
20 * Redistributions in binary form must reproduce the above copyright notice,
21 this list of conditions and the following disclaimer in the documentation
22 and/or other materials provided with the distribution.
24 * Neither the name of the Florida State University nor the names of its
25 contributors may be used to endorse or promote products derived from this
26 software without specific prior written permission.
28 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
29 ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
30 WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
31 DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
32 ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
33 (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
34 LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
35 ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
36 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
37 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
39 *************************************************************************
40 * Any results obtained from executing this software require the *
41 * acknowledgment and citation of the software and its owners. *
42 * The full citation is given below: *
44 * A. Faraj and X. Yuan. "Automatic Generation and Tuning of MPI *
45 * Collective Communication Routines." The 19th ACM International *
46 * Conference on Supercomputing (ICS), Cambridge, Massachusetts, *
48 *************************************************************************
50 *****************************************************************************/
52 /*****************************************************************************
55 * num: the number of processors in a communicator
59 * descp: takes a number and tries to find a factoring of x*y*z mesh out of it
60 ****************************************************************************/
63 static int is_3dmesh(int num, int *i, int *j, int *k)
69 if ((num % (x * x)) == 0) {
79 /*****************************************************************************
80 * Function: allgather_3dmesh_shoot
82 * send_buff: send input buffer
83 * send_count: number of elements to send
84 * send_type: data type of elements being sent
85 * recv_buff: receive output buffer
86 * recv_count: number of elements to received
87 * recv_type: data type of elements being received
89 * Descrp: Function realizes the allgather operation using the 2dmesh
90 * algorithm. Allgather ommunication occurs first in the x dimension, y
91 * dimension, and then in the z dimension. Communication in each dimension
94 ****************************************************************************/
99 int Coll_allgather_3dmesh::allgather(const void *send_buff, int send_count,
100 MPI_Datatype send_type, void *recv_buff,
101 int recv_count, MPI_Datatype recv_type,
106 int i, src, dst, rank, num_procs, block_size, my_z_base;
107 int my_z, X, Y, Z, send_offset, recv_offset;
108 int two_dsize, my_row_base, my_col_base, src_row_base, src_z_base, num_reqs;
109 int tag = COLL_TAG_ALLGATHER;
112 num_procs = comm->size();
113 extent = send_type->get_extent();
115 if (not is_3dmesh(num_procs, &X, &Y, &Z))
116 throw std::invalid_argument("allgather_3dmesh algorithm can't be used with this number of processes!");
126 my_z = rank / two_dsize;
128 my_row_base = (rank / X) * X;
129 my_col_base = (rank % Y) + (my_z * two_dsize);
130 my_z_base = my_z * two_dsize;
132 block_size = extent * send_count;
134 MPI_Request* req = new MPI_Request[num_reqs];
135 MPI_Request* req_ptr = req;
137 // do local allgather/local copy
138 recv_offset = rank * block_size;
139 Datatype::copy(send_buff, send_count, send_type, (char *)recv_buff + recv_offset,
140 recv_count, recv_type);
143 for (i = 0; i < Y; i++) {
144 src = i + my_row_base;
147 recv_offset = src * block_size;
148 *(req_ptr++) = Request::irecv((char *)recv_buff + recv_offset, send_count, recv_type, src, tag,
152 for (i = 0; i < Y; i++) {
153 dst = i + my_row_base;
156 Request::send(send_buff, send_count, send_type, dst, tag, comm);
159 Request::waitall(Y - 1, req, MPI_STATUSES_IGNORE);
162 // do colwise comm, it does not matter here if i*X or i *Y since X == Y
164 for (i = 0; i < X; i++) {
165 src = (i * Y + my_col_base);
169 src_row_base = (src / X) * X;
170 recv_offset = src_row_base * block_size;
171 *(req_ptr++) = Request::irecv((char *)recv_buff + recv_offset, recv_count * Y, recv_type, src, tag,
175 send_offset = my_row_base * block_size;
177 for (i = 0; i < X; i++) {
178 dst = (i * Y + my_col_base);
181 Request::send((char *)recv_buff + send_offset, send_count * Y, send_type, dst, tag,
185 Request::waitall(X - 1, req, MPI_STATUSES_IGNORE);
188 for (i = 1; i < Z; i++) {
189 src = (rank + i * two_dsize) % num_procs;
190 src_z_base = (src / two_dsize) * two_dsize;
192 recv_offset = (src_z_base * block_size);
194 *(req_ptr++) = Request::irecv((char *)recv_buff + recv_offset, recv_count * two_dsize, recv_type,
198 for (i = 1; i < Z; i++) {
199 dst = (rank + i * two_dsize) % num_procs;
200 send_offset = my_z_base * block_size;
201 Request::send((char *)recv_buff + send_offset, send_count * two_dsize, send_type,
204 Request::waitall(Z - 1, req, MPI_STATUSES_IGNORE);