2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 1991-2000, University of Groningen, The Netherlands.
5 * Copyright (c) 2001-2004, The GROMACS development team.
6 * Copyright (c) 2013,2014,2015,2016,2017,2018,2019, by the GROMACS development team, led by
7 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
8 * and including many others, as listed in the AUTHORS file in the
9 * top-level source directory and at http://www.gromacs.org.
11 * GROMACS is free software; you can redistribute it and/or
12 * modify it under the terms of the GNU Lesser General Public License
13 * as published by the Free Software Foundation; either version 2.1
14 * of the License, or (at your option) any later version.
16 * GROMACS is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 * Lesser General Public License for more details.
21 * You should have received a copy of the GNU Lesser General Public
22 * License along with GROMACS; if not, see
23 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
24 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
26 * If you want to redistribute modifications to GROMACS, please
27 * consider that scientific software is very special. Version
28 * control is crucial - bugs must be traceable. We will be happy to
29 * consider code for inclusion in the official distribution, but
30 * derived work must not be called official GROMACS. Details are found
31 * in the README & COPYING files - if they are missing, get the
32 * official version at http://www.gromacs.org.
34 * To help us fund GROMACS development, we humbly ask that you cite
35 * the research papers on the package. Check out http://www.gromacs.org.
48 #include "gromacs/commandline/filenm.h"
49 #include "gromacs/mdtypes/commrec.h"
50 #include "gromacs/utility/basenetwork.h"
51 #include "gromacs/utility/cstringutil.h"
52 #include "gromacs/utility/fatalerror.h"
53 #include "gromacs/utility/futil.h"
54 #include "gromacs/utility/gmxmpi.h"
55 #include "gromacs/utility/mpiinplacebuffers.h"
56 #include "gromacs/utility/real.h"
57 #include "gromacs/utility/smalloc.h"
59 /* The source code in this file should be thread-safe.
60 Please keep it that way. */
62 void gmx_fill_commrec_from_mpi(t_commrec *cr)
65 gmx_call("gmx_fill_commrec_from_mpi");
68 if (!gmx_mpi_initialized())
70 gmx_comm("MPI has not been initialized properly");
73 cr->nnodes = gmx_node_num();
74 cr->nodeid = gmx_node_rank();
75 cr->sim_nodeid = cr->nodeid;
76 cr->mpi_comm_mysim = MPI_COMM_WORLD;
77 cr->mpi_comm_mygroup = MPI_COMM_WORLD;
81 t_commrec *init_commrec()
88 gmx_fill_commrec_from_mpi(cr);
90 cr->mpi_comm_mysim = MPI_COMM_NULL;
91 cr->mpi_comm_mygroup = MPI_COMM_NULL;
94 cr->nodeid = cr->sim_nodeid;
97 // TODO cr->duty should not be initialized here
98 cr->duty = (DUTY_PP | DUTY_PME);
100 #if GMX_MPI && !MPI_IN_PLACE_EXISTS
101 /* initialize the MPI_IN_PLACE replacement buffers */
103 cr->mpb->ibuf = nullptr;
104 cr->mpb->libuf = nullptr;
105 cr->mpb->fbuf = nullptr;
106 cr->mpb->dbuf = nullptr;
107 cr->mpb->ibuf_alloc = 0;
108 cr->mpb->libuf_alloc = 0;
109 cr->mpb->fbuf_alloc = 0;
110 cr->mpb->dbuf_alloc = 0;
116 void done_commrec(t_commrec *cr)
120 if (nullptr != cr->dd)
123 // done_domdec(cr->dd);
125 done_mpi_in_place_buf(cr->mpb);
128 // TODO We need to be able to free communicators, but the
129 // structure of the commrec and domdec initialization code makes
130 // it hard to avoid both leaks and double frees.
131 bool mySimIsMyGroup = (cr->mpi_comm_mysim == cr->mpi_comm_mygroup);
132 if (cr->mpi_comm_mysim != MPI_COMM_NULL &&
133 cr->mpi_comm_mysim != MPI_COMM_WORLD)
136 // MPI_Comm_free(&cr->mpi_comm_mysim);
138 if (!mySimIsMyGroup &&
139 cr->mpi_comm_mygroup != MPI_COMM_NULL &&
140 cr->mpi_comm_mygroup != MPI_COMM_WORLD)
143 // MPI_Comm_free(&cr->mpi_comm_mygroup);
149 t_commrec *reinitialize_commrec_for_this_thread(const t_commrec *cro)
154 /* make a thread-specific commrec */
156 /* now copy the whole thing, so settings like the number of PME nodes
160 /* and we start setting our own thread-specific values for things */
161 gmx_fill_commrec_from_mpi(cr);
163 // TODO cr->duty should not be initialized here
164 cr->duty = (DUTY_PP | DUTY_PME);
168 GMX_UNUSED_VALUE(cro);
173 void gmx_setup_nodecomm(FILE gmx_unused *fplog, t_commrec *cr)
177 /* Many MPI implementations do not optimize MPI_Allreduce
178 * (and probably also other global communication calls)
179 * for multi-core nodes connected by a network.
180 * We can optimize such communication by using one MPI call
181 * within each node and one between the nodes.
182 * For MVAPICH2 and Intel MPI this reduces the time for
183 * the global_stat communication by 25%
184 * for 2x2-core 3 GHz Woodcrest connected by mixed DDR/SDR Infiniband.
185 * B. Hess, November 2007
195 // TODO PhysicalNodeCommunicator could be extended/used to handle
196 // the need for per-node per-group communicators.
197 MPI_Comm_size(cr->mpi_comm_mygroup, &n);
198 MPI_Comm_rank(cr->mpi_comm_mygroup, &rank);
200 int nodehash = gmx_physicalnode_id_hash();
204 fprintf(debug, "In gmx_setup_nodecomm: splitting communicator of size %d\n", n);
208 /* The intra-node communicator, split on node number */
209 MPI_Comm_split(cr->mpi_comm_mygroup, nodehash, rank, &nc->comm_intra);
210 MPI_Comm_rank(nc->comm_intra, &nc->rank_intra);
213 fprintf(debug, "In gmx_setup_nodecomm: node ID %d rank within node %d\n",
214 rank, nc->rank_intra);
216 /* The inter-node communicator, split on rank_intra.
217 * We actually only need the one for rank=0,
218 * but it is easier to create them all.
220 MPI_Comm_split(cr->mpi_comm_mygroup, nc->rank_intra, rank, &nc->comm_inter);
221 /* Check if this really created two step communication */
224 MPI_Comm_size(nc->comm_inter, &ng);
225 MPI_Comm_size(nc->comm_intra, &ni);
228 fprintf(debug, "In gmx_setup_nodecomm: groups %d, my group size %d\n",
232 if (getenv("GMX_NO_NODECOMM") == nullptr &&
233 ((ng > 1 && ng < n) || (ni > 1 && ni < n)))
238 fprintf(fplog, "Using two step summing over %d groups of on average %.1f ranks\n\n",
239 ng, (real)n/(real)ng);
241 if (nc->rank_intra > 0)
243 MPI_Comm_free(&nc->comm_inter);
248 /* One group or all processes in a separate group, use normal summing */
249 MPI_Comm_free(&nc->comm_inter);
250 MPI_Comm_free(&nc->comm_intra);
253 fprintf(debug, "In gmx_setup_nodecomm: not unsing separate inter- and intra-node communicators.\n");
258 /* tMPI runs only on a single node so just use the nodeid */
259 nc->rank_intra = cr->nodeid;
263 void gmx_barrier(const t_commrec gmx_unused *cr)
266 gmx_call("gmx_barrier");
268 MPI_Barrier(cr->mpi_comm_mygroup);
272 void gmx_bcast(int gmx_unused nbytes, void gmx_unused *b, const t_commrec gmx_unused *cr)
275 gmx_call("gmx_bast");
277 MPI_Bcast(b, nbytes, MPI_BYTE, MASTERRANK(cr), cr->mpi_comm_mygroup);
281 void gmx_bcast_sim(int gmx_unused nbytes, void gmx_unused *b, const t_commrec gmx_unused *cr)
284 gmx_call("gmx_bast");
286 MPI_Bcast(b, nbytes, MPI_BYTE, MASTERRANK(cr), cr->mpi_comm_mysim);
290 void gmx_sumd(int gmx_unused nr, double gmx_unused r[], const t_commrec gmx_unused *cr)
293 gmx_call("gmx_sumd");
295 #if MPI_IN_PLACE_EXISTS
298 if (cr->nc.rank_intra == 0)
300 /* Use two step summing. */
301 MPI_Reduce(MPI_IN_PLACE, r, nr, MPI_DOUBLE, MPI_SUM, 0,
303 /* Sum the roots of the internal (intra) buffers. */
304 MPI_Allreduce(MPI_IN_PLACE, r, nr, MPI_DOUBLE, MPI_SUM,
309 /* This is here because of the silly MPI specification
310 that MPI_IN_PLACE should be put in sendbuf instead of recvbuf */
311 MPI_Reduce(r, nullptr, nr, MPI_DOUBLE, MPI_SUM, 0, cr->nc.comm_intra);
313 MPI_Bcast(r, nr, MPI_DOUBLE, 0, cr->nc.comm_intra);
317 MPI_Allreduce(MPI_IN_PLACE, r, nr, MPI_DOUBLE, MPI_SUM,
318 cr->mpi_comm_mygroup);
323 if (nr > cr->mpb->dbuf_alloc)
325 cr->mpb->dbuf_alloc = nr;
326 srenew(cr->mpb->dbuf, cr->mpb->dbuf_alloc);
330 /* Use two step summing */
331 MPI_Allreduce(r, cr->mpb->dbuf, nr, MPI_DOUBLE, MPI_SUM, cr->nc.comm_intra);
332 if (cr->nc.rank_intra == 0)
334 /* Sum with the buffers reversed */
335 MPI_Allreduce(cr->mpb->dbuf, r, nr, MPI_DOUBLE, MPI_SUM,
338 MPI_Bcast(r, nr, MPI_DOUBLE, 0, cr->nc.comm_intra);
342 MPI_Allreduce(r, cr->mpb->dbuf, nr, MPI_DOUBLE, MPI_SUM,
343 cr->mpi_comm_mygroup);
344 for (i = 0; i < nr; i++)
346 r[i] = cr->mpb->dbuf[i];
353 void gmx_sumf(int gmx_unused nr, float gmx_unused r[], const t_commrec gmx_unused *cr)
356 gmx_call("gmx_sumf");
358 #if MPI_IN_PLACE_EXISTS
361 /* Use two step summing. */
362 if (cr->nc.rank_intra == 0)
364 MPI_Reduce(MPI_IN_PLACE, r, nr, MPI_FLOAT, MPI_SUM, 0,
366 /* Sum the roots of the internal (intra) buffers */
367 MPI_Allreduce(MPI_IN_PLACE, r, nr, MPI_FLOAT, MPI_SUM,
372 /* This is here because of the silly MPI specification
373 that MPI_IN_PLACE should be put in sendbuf instead of recvbuf */
374 MPI_Reduce(r, nullptr, nr, MPI_FLOAT, MPI_SUM, 0, cr->nc.comm_intra);
376 MPI_Bcast(r, nr, MPI_FLOAT, 0, cr->nc.comm_intra);
380 MPI_Allreduce(MPI_IN_PLACE, r, nr, MPI_FLOAT, MPI_SUM, cr->mpi_comm_mygroup);
385 if (nr > cr->mpb->fbuf_alloc)
387 cr->mpb->fbuf_alloc = nr;
388 srenew(cr->mpb->fbuf, cr->mpb->fbuf_alloc);
392 /* Use two step summing */
393 MPI_Allreduce(r, cr->mpb->fbuf, nr, MPI_FLOAT, MPI_SUM, cr->nc.comm_intra);
394 if (cr->nc.rank_intra == 0)
396 /* Sum with the buffers reversed */
397 MPI_Allreduce(cr->mpb->fbuf, r, nr, MPI_FLOAT, MPI_SUM,
400 MPI_Bcast(r, nr, MPI_FLOAT, 0, cr->nc.comm_intra);
404 MPI_Allreduce(r, cr->mpb->fbuf, nr, MPI_FLOAT, MPI_SUM,
405 cr->mpi_comm_mygroup);
406 for (i = 0; i < nr; i++)
408 r[i] = cr->mpb->fbuf[i];
415 void gmx_sumi(int gmx_unused nr, int gmx_unused r[], const t_commrec gmx_unused *cr)
418 gmx_call("gmx_sumi");
420 #if MPI_IN_PLACE_EXISTS
423 /* Use two step summing */
424 if (cr->nc.rank_intra == 0)
426 MPI_Reduce(MPI_IN_PLACE, r, nr, MPI_INT, MPI_SUM, 0, cr->nc.comm_intra);
427 /* Sum with the buffers reversed */
428 MPI_Allreduce(MPI_IN_PLACE, r, nr, MPI_INT, MPI_SUM, cr->nc.comm_inter);
432 /* This is here because of the silly MPI specification
433 that MPI_IN_PLACE should be put in sendbuf instead of recvbuf */
434 MPI_Reduce(r, nullptr, nr, MPI_INT, MPI_SUM, 0, cr->nc.comm_intra);
436 MPI_Bcast(r, nr, MPI_INT, 0, cr->nc.comm_intra);
440 MPI_Allreduce(MPI_IN_PLACE, r, nr, MPI_INT, MPI_SUM, cr->mpi_comm_mygroup);
445 if (nr > cr->mpb->ibuf_alloc)
447 cr->mpb->ibuf_alloc = nr;
448 srenew(cr->mpb->ibuf, cr->mpb->ibuf_alloc);
452 /* Use two step summing */
453 MPI_Allreduce(r, cr->mpb->ibuf, nr, MPI_INT, MPI_SUM, cr->nc.comm_intra);
454 if (cr->nc.rank_intra == 0)
456 /* Sum with the buffers reversed */
457 MPI_Allreduce(cr->mpb->ibuf, r, nr, MPI_INT, MPI_SUM, cr->nc.comm_inter);
459 MPI_Bcast(r, nr, MPI_INT, 0, cr->nc.comm_intra);
463 MPI_Allreduce(r, cr->mpb->ibuf, nr, MPI_INT, MPI_SUM, cr->mpi_comm_mygroup);
464 for (i = 0; i < nr; i++)
466 r[i] = cr->mpb->ibuf[i];
473 void gmx_sumli(int gmx_unused nr, int64_t gmx_unused r[], const t_commrec gmx_unused *cr)
476 gmx_call("gmx_sumli");
478 #if MPI_IN_PLACE_EXISTS
481 /* Use two step summing */
482 if (cr->nc.rank_intra == 0)
484 MPI_Reduce(MPI_IN_PLACE, r, nr, MPI_INT64_T, MPI_SUM, 0,
486 /* Sum with the buffers reversed */
487 MPI_Allreduce(MPI_IN_PLACE, r, nr, MPI_INT64_T, MPI_SUM,
492 /* This is here because of the silly MPI specification
493 that MPI_IN_PLACE should be put in sendbuf instead of recvbuf */
494 MPI_Reduce(r, nullptr, nr, MPI_INT64_T, MPI_SUM, 0, cr->nc.comm_intra);
496 MPI_Bcast(r, nr, MPI_INT64_T, 0, cr->nc.comm_intra);
500 MPI_Allreduce(MPI_IN_PLACE, r, nr, MPI_INT64_T, MPI_SUM, cr->mpi_comm_mygroup);
505 if (nr > cr->mpb->libuf_alloc)
507 cr->mpb->libuf_alloc = nr;
508 srenew(cr->mpb->libuf, cr->mpb->libuf_alloc);
512 /* Use two step summing */
513 MPI_Allreduce(r, cr->mpb->libuf, nr, MPI_INT64_T, MPI_SUM,
515 if (cr->nc.rank_intra == 0)
517 /* Sum with the buffers reversed */
518 MPI_Allreduce(cr->mpb->libuf, r, nr, MPI_INT64_T, MPI_SUM,
521 MPI_Bcast(r, nr, MPI_INT64_T, 0, cr->nc.comm_intra);
525 MPI_Allreduce(r, cr->mpb->libuf, nr, MPI_INT64_T, MPI_SUM,
526 cr->mpi_comm_mygroup);
527 for (i = 0; i < nr; i++)
529 r[i] = cr->mpb->libuf[i];
536 const char *opt2fn_master(const char *opt, int nfile, const t_filenm fnm[],
539 return SIMMASTER(cr) ? opt2fn(opt, nfile, fnm) : nullptr;
542 void gmx_fatal_collective(int f_errno, const char *file, int line,
543 MPI_Comm comm, gmx_bool bMaster,
544 gmx_fmtstr const char *fmt, ...)
550 /* Check if we are calling on all processes in MPI_COMM_WORLD */
551 MPI_Comm_compare(comm, MPI_COMM_WORLD, &result);
552 /* Any result except MPI_UNEQUAL allows us to call MPI_Finalize */
553 bFinalize = (result != MPI_UNEQUAL);
555 GMX_UNUSED_VALUE(comm);
560 gmx_fatal_mpi_va(f_errno, file, line, bMaster, bFinalize, fmt, ap);