Bug Summary

File:gromacs/gmxlib/gmx_thread_affinity.c
Location:line 292, column 5
Description:Value stored to 'nth_affinity_set' is never read

Annotated Source Code

1/*
2 * This file is part of the GROMACS molecular simulation package.
3 *
4 * Copyright (c) 2012,2013,2014, by the GROMACS development team, led by
5 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
6 * and including many others, as listed in the AUTHORS file in the
7 * top-level source directory and at http://www.gromacs.org.
8 *
9 * GROMACS is free software; you can redistribute it and/or
10 * modify it under the terms of the GNU Lesser General Public License
11 * as published by the Free Software Foundation; either version 2.1
12 * of the License, or (at your option) any later version.
13 *
14 * GROMACS is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 * Lesser General Public License for more details.
18 *
19 * You should have received a copy of the GNU Lesser General Public
20 * License along with GROMACS; if not, see
21 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
22 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
23 *
24 * If you want to redistribute modifications to GROMACS, please
25 * consider that scientific software is very special. Version
26 * control is crucial - bugs must be traceable. We will be happy to
27 * consider code for inclusion in the official distribution, but
28 * derived work must not be called official GROMACS. Details are found
29 * in the README & COPYING files - if they are missing, get the
30 * official version at http://www.gromacs.org.
31 *
32 * To help us fund GROMACS development, we humbly ask that you cite
33 * the research papers on the package. Check out http://www.gromacs.org.
34 */
35#ifdef HAVE_CONFIG_H1
36#include <config.h>
37#endif
38#if defined(HAVE_SCHED_H) && defined(HAVE_SCHED_GETAFFINITY)
39#define _GNU_SOURCE
40#include <sched.h>
41#include <sys/syscall.h>
42#endif
43#include <string.h>
44#include <errno(*__errno_location ()).h>
45#include <assert.h>
46#include <stdio.h>
47
48#include "thread_mpi/threads.h"
49
50#include "typedefs.h"
51#include "types/commrec.h"
52#include "types/hw_info.h"
53#include "copyrite.h"
54#include "gmx_cpuid.h"
55#include "gmx_omp_nthreads.h"
56#include "md_logging.h"
57#include "gmx_thread_affinity.h"
58
59#include "gromacs/utility/basenetwork.h"
60#include "gromacs/utility/cstringutil.h"
61#include "gromacs/utility/fatalerror.h"
62#include "gromacs/utility/gmxomp.h"
63#include "gromacs/utility/smalloc.h"
64
65static int
66get_thread_affinity_layout(FILE *fplog,
67 const t_commrec *cr,
68 const gmx_hw_info_t * hwinfo,
69 int nthreads,
70 int pin_offset, int * pin_stride,
71 const int **locality_order)
72{
73 int nhwthreads, npkg, ncores, nhwthreads_per_core, rc;
74 const int * pkg_id;
75 const int * core_id;
76 const int * hwthread_id;
77 gmx_bool bPickPinStride;
78
79 if (pin_offset < 0)
80 {
81 gmx_fatal(FARGS0, "/home/alexxy/Develop/gromacs/src/gromacs/gmxlib/gmx_thread_affinity.c"
, 81, "Negative thread pinning offset requested");
82 }
83 if (*pin_stride < 0)
84 {
85 gmx_fatal(FARGS0, "/home/alexxy/Develop/gromacs/src/gromacs/gmxlib/gmx_thread_affinity.c"
, 85, "Negative thread pinning stride requested");
86 }
87
88 rc = gmx_cpuid_topology(hwinfo->cpuid_info, &nhwthreads, &npkg, &ncores,
89 &nhwthreads_per_core,
90 &pkg_id, &core_id, &hwthread_id, locality_order);
91
92 if (rc != 0)
93 {
94 /* topology information not available or invalid, ignore it */
95 nhwthreads = hwinfo->nthreads_hw_avail;
96 *locality_order = NULL((void*)0);
97
98 if (nhwthreads <= 0)
99 {
100 /* We don't know anything about the hardware, don't pin */
101 md_print_warn(cr, fplog,
102 "NOTE: We don't know how many logical cores we have, will not pin threads");
103
104 return -1;
105 }
106 }
107
108 if (nthreads > nhwthreads)
109 {
110 /* We are oversubscribing, don't pin */
111 md_print_warn(NULL((void*)0), fplog,
112 "WARNING: Oversubscribing the CPU, will not pin threads");
113
114 return -1;
115 }
116
117 if (pin_offset + nthreads > nhwthreads)
118 {
119 /* We are oversubscribing, don't pin */
120 md_print_warn(NULL((void*)0), fplog,
121 "WARNING: The requested pin offset is too large for the available logical cores,\n"
122 " will not pin threads");
123
124 return -1;
125 }
126
127
128 /* do we need to choose the pinning stride? */
129 bPickPinStride = (*pin_stride == 0);
130
131 if (bPickPinStride)
132 {
133 if (rc == 0 && pin_offset + nthreads*nhwthreads_per_core <= nhwthreads)
134 {
135 /* Put one thread on each physical core */
136 *pin_stride = nhwthreads_per_core;
137 }
138 else
139 {
140 /* We don't know if we have SMT, and if we do, we don't know
141 * if hw threads in the same physical core are consecutive.
142 * Without SMT the pinning layout should not matter too much.
143 * so we assume a consecutive layout and maximally spread out"
144 * the threads at equal threads per core.
145 * Note that IBM is the major non-x86 case with cpuid support
146 * and probably threads are already pinned by the queuing system,
147 * so we wouldn't end up here in the first place.
148 */
149 *pin_stride = (nhwthreads - pin_offset)/nthreads;
150 }
151 }
152 else
153 {
154 /* Check the placement of the thread with the largest index to make sure
155 * that the offset & stride doesn't cause pinning beyond the last hardware thread. */
156 if (pin_offset + (nthreads-1)*(*pin_stride) >= nhwthreads)
157 {
158 /* We are oversubscribing, don't pin */
159 md_print_warn(NULL((void*)0), fplog,
160 "WARNING: The requested pinning stride is too large for the available logical cores,\n"
161 " will not pin threads");
162
163 return -1;
164 }
165 }
166
167 if (fplog != NULL((void*)0))
168 {
169 fprintf(fplog, "Pinning threads with a%s logical core stride of %d\n",
170 bPickPinStride ? "n auto-selected" : " user-specified",
171 *pin_stride);
172 }
173
174 return 0;
175}
176
177/* Set CPU affinity. Can be important for performance.
178 On some systems (e.g. Cray) CPU Affinity is set by default.
179 But default assigning doesn't work (well) with only some ranks
180 having threads. This causes very low performance.
181 External tools have cumbersome syntax for setting affinity
182 in the case that only some ranks have threads.
183 Thus it is important that GROMACS sets the affinity internally
184 if only PME is using threads.
185 */
186void
187gmx_set_thread_affinity(FILE *fplog,
188 const t_commrec *cr,
189 gmx_hw_opt_t *hw_opt,
190 const gmx_hw_info_t *hwinfo)
191{
192 int nth_affinity_set, thread0_id_node,
193 nthread_local, nthread_node, nthread_hw_max, nphyscore;
194 int offset;
195 const int *locality_order;
196 int rc;
197
198 if (hw_opt->thread_affinity == threadaffOFF)
199 {
200 /* Nothing to do */
201 return;
202 }
203
204 /* If the tMPI thread affinity setting is not supported encourage the user
205 * to report it as it's either a bug or an exotic platform which we might
206 * want to support. */
207 if (tMPI_Thread_setaffinity_support() != TMPI_SETAFFINITY_SUPPORT_YES)
208 {
209 /* we know Mac OS doesn't support setting thread affinity, so there's
210 no point in warning the user in that case. In any other case
211 the user might be able to do something about it. */
212#ifndef __APPLE__
213 md_print_warn(NULL((void*)0), fplog,
214 "Can not set thread affinities on the current platform. On NUMA systems this\n"
215 "can cause performance degradation. If you think your platform should support\n"
216 "setting affinities, contact the GROMACS developers.");
217#endif /* __APPLE__ */
218 return;
219 }
220
221 /* threads on this MPI process or TMPI thread */
222 if (cr->duty & DUTY_PP(1<<0))
223 {
224 nthread_local = gmx_omp_nthreads_get(emntNonbonded);
225 }
226 else
227 {
228 nthread_local = gmx_omp_nthreads_get(emntPME);
229 }
230
231 /* map the current process to cores */
232 thread0_id_node = 0;
233 nthread_node = nthread_local;
234#ifdef GMX_MPI
235 if (PAR(cr)((cr)->nnodes > 1) || MULTISIM(cr)((cr)->ms))
236 {
237 /* We need to determine a scan of the thread counts in this
238 * compute node.
239 */
240 MPI_Comm comm_intra;
241
242 MPI_Comm_splittMPI_Comm_split(MPI_COMM_WORLDTMPI_COMM_WORLD, gmx_hostname_num(), cr->rank_intranode,
243 &comm_intra);
244 MPI_ScantMPI_Scan(&nthread_local, &thread0_id_node, 1, MPI_INTTMPI_INT, MPI_SUMTMPI_SUM, comm_intra);
245 /* MPI_Scan is inclusive, but here we need exclusive */
246 thread0_id_node -= nthread_local;
247 /* Get the total number of threads on this physical node */
248 MPI_AllreducetMPI_Allreduce(&nthread_local, &nthread_node, 1, MPI_INTTMPI_INT, MPI_SUMTMPI_SUM, comm_intra);
249 MPI_Comm_freetMPI_Comm_free(&comm_intra);
250 }
251#endif
252
253 if (hw_opt->thread_affinity == threadaffAUTO &&
254 nthread_node != hwinfo->nthreads_hw_avail)
255 {
256 if (nthread_node > 1 && nthread_node < hwinfo->nthreads_hw_avail)
257 {
258 md_print_warn(cr, fplog,
259 "NOTE: The number of threads is not equal to the number of (logical) cores\n"
260 " and the -pin option is set to auto: will not pin thread to cores.\n"
261 " This can lead to significant performance degradation.\n"
262 " Consider using -pin on (and -pinoffset in case you run multiple jobs).\n");
263 }
264
265 return;
266 }
267
268 offset = 0;
269 if (hw_opt->core_pinning_offset != 0)
270 {
271 offset = hw_opt->core_pinning_offset;
272 md_print_info(cr, fplog, "Applying core pinning offset %d\n", offset);
273 }
274
275 rc = get_thread_affinity_layout(fplog, cr, hwinfo,
276 nthread_node,
277 offset, &hw_opt->core_pinning_stride,
278 &locality_order);
279
280 if (rc != 0)
281 {
282 /* Incompatible layout, don't pin, warning was already issued */
283 return;
284 }
285
286 /* Set the per-thread affinity. In order to be able to check the success
287 * of affinity settings, we will set nth_affinity_set to 1 on threads
288 * where the affinity setting succeded and to 0 where it failed.
289 * Reducing these 0/1 values over the threads will give the total number
290 * of threads on which we succeeded.
291 */
292 nth_affinity_set = 0;
Value stored to 'nth_affinity_set' is never read
293#pragma omp parallel num_threads(nthread_local) reduction(+:nth_affinity_set)
294 {
295 int thread_id, thread_id_node;
296 int index, core;
297 gmx_bool setaffinity_ret;
298
299 thread_id = gmx_omp_get_thread_num();
300 thread_id_node = thread0_id_node + thread_id;
301 index = offset + thread_id_node*hw_opt->core_pinning_stride;
302 if (locality_order != NULL((void*)0))
303 {
304 core = locality_order[index];
305 }
306 else
307 {
308 core = index;
309 }
310
311 setaffinity_ret = tMPI_Thread_setaffinity_single(tMPI_Thread_self(), core);
312
313 /* store the per-thread success-values of the setaffinity */
314 nth_affinity_set = (setaffinity_ret == 0);
315
316 if (debug)
317 {
318 fprintf(debug, "On rank %2d, thread %2d, index %2d, core %2d the affinity setting returned %d\n",
319 cr->nodeid, gmx_omp_get_thread_num(), index, core, setaffinity_ret);
320 }
321 }
322
323 if (nth_affinity_set > nthread_local)
324 {
325 char msg[STRLEN4096];
326
327 sprintf(msg, "Looks like we have set affinity for more threads than "
328 "we have (%d > %d)!\n", nth_affinity_set, nthread_local);
329 gmx_incons(msg)_gmx_error("incons", msg, "/home/alexxy/Develop/gromacs/src/gromacs/gmxlib/gmx_thread_affinity.c"
, 329);
330 }
331 else
332 {
333 /* check & warn if some threads failed to set their affinities */
334 if (nth_affinity_set != nthread_local)
335 {
336 char sbuf1[STRLEN4096], sbuf2[STRLEN4096];
337
338 /* sbuf1 contains rank info, while sbuf2 OpenMP thread info */
339 sbuf1[0] = sbuf2[0] = '\0';
340 /* Only add rank info if we have more than one rank. */
341 if (cr->nnodes > 1)
342 {
343#ifdef GMX_MPI
344#ifdef GMX_THREAD_MPI
345 sprintf(sbuf1, "In tMPI thread #%d: ", cr->nodeid);
346#else /* GMX_LIB_MPI */
347 sprintf(sbuf1, "In MPI process #%d: ", cr->nodeid);
348#endif
349#endif /* GMX_MPI */
350 }
351
352 if (nthread_local > 1)
353 {
354 sprintf(sbuf2, "for %d/%d thread%s ",
355 nthread_local - nth_affinity_set, nthread_local,
356 nthread_local > 1 ? "s" : "");
357 }
358
359 md_print_warn(NULL((void*)0), fplog,
360 "WARNING: %sAffinity setting %sfailed.\n"
361 " This can cause performance degradation! If you think your setting are\n"
362 " correct, contact the GROMACS developers.",
363 sbuf1, sbuf2);
364 }
365 }
366 return;
367}
368
369/* Check the process affinity mask and if it is found to be non-zero,
370 * will honor it and disable mdrun internal affinity setting.
371 * Note that this will only work on Linux as we use a GNU feature.
372 */
373void
374gmx_check_thread_affinity_set(FILE *fplog,
375 const t_commrec *cr,
376 gmx_hw_opt_t *hw_opt,
377 int gmx_unused__attribute__ ((unused)) ncpus,
378 gmx_bool bAfterOpenmpInit)
379{
380#ifdef HAVE_SCHED_GETAFFINITY
381 cpu_set_t mask_current;
382 int i, ret, cpu_count, cpu_set;
383 gmx_bool bAllSet;
384#endif
385
386 assert(hw_opt)((void) (0));
387 if (!bAfterOpenmpInit)
388 {
389 /* Check for externally set OpenMP affinity and turn off internal
390 * pinning if any is found. We need to do this check early to tell
391 * thread-MPI whether it should do pinning when spawning threads.
392 * TODO: the above no longer holds, we should move these checks later
393 */
394 if (hw_opt->thread_affinity != threadaffOFF)
395 {
396 char *message;
397 if (!gmx_omp_check_thread_affinity(&message))
398 {
399 /* TODO: with -pin auto we should only warn when using all cores */
400 md_print_warn(cr, fplog, "%s", message);
401 sfree(message)save_free("message", "/home/alexxy/Develop/gromacs/src/gromacs/gmxlib/gmx_thread_affinity.c"
, 401, (message));
402 hw_opt->thread_affinity = threadaffOFF;
403 }
404 }
405
406 /* With thread-MPI this is needed as pinning might get turned off,
407 * which needs to be known before starting thread-MPI.
408 * With thread-MPI hw_opt is processed here on the master rank
409 * and passed to the other ranks later, so we only do this on master.
410 */
411 if (!SIMMASTER(cr)(((((cr)->nodeid == 0) || !((cr)->nnodes > 1)) &&
((cr)->duty & (1<<0))) || !((cr)->nnodes >
1)))
412 {
413 return;
414 }
415#ifndef GMX_THREAD_MPI
416 return;
417#endif
418 }
419
420#ifdef HAVE_SCHED_GETAFFINITY
421 if (hw_opt->thread_affinity == threadaffOFF)
422 {
423 /* internal affinity setting is off, don't bother checking process affinity */
424 return;
425 }
426
427 CPU_ZERO(&mask_current)do __builtin_memset (&mask_current, '\0', sizeof (cpu_set_t
)); while (0);
428 if ((ret = sched_getaffinity(0, sizeof(cpu_set_t), &mask_current)) != 0)
429 {
430 /* failed to query affinity mask, will just return */
431 if (debug)
432 {
433 fprintf(debug, "Failed to query affinity mask (error %d)", ret);
434 }
435 return;
436 }
437
438 /* Before proceeding with the actual check, make sure that the number of
439 * detected CPUs is >= the CPUs in the current set.
440 * We need to check for CPU_COUNT as it was added only in glibc 2.6. */
441#ifdef CPU_COUNT
442 if (ncpus < CPU_COUNT(&mask_current)__sched_cpucount (sizeof (cpu_set_t), &mask_current))
443 {
444 if (debug)
445 {
446 fprintf(debug, "%d CPUs detected, but %d was returned by CPU_COUNT",
447 ncpus, CPU_COUNT(&mask_current)__sched_cpucount (sizeof (cpu_set_t), &mask_current));
448 }
449 return;
450 }
451#endif /* CPU_COUNT */
452
453 bAllSet = TRUE1;
454 for (i = 0; (i < ncpus && i < CPU_SETSIZE1024); i++)
455 {
456 bAllSet = bAllSet && (CPU_ISSET(i, &mask_current)(__extension__ ({ size_t __cpu = (i); __cpu / 8 < (sizeof (
cpu_set_t)) ? ((((const __cpu_mask *) ((&mask_current)->
__bits))[((__cpu) / (8 * sizeof (__cpu_mask)))] & ((__cpu_mask
) 1 << ((__cpu) % (8 * sizeof (__cpu_mask)))))) != 0 : 0
; })) != 0);
457 }
458
459 if (!bAllSet)
460 {
461 if (hw_opt->thread_affinity == threadaffAUTO)
462 {
463 if (!bAfterOpenmpInit)
464 {
465 md_print_warn(cr, fplog,
466 "Non-default thread affinity set, disabling internal thread affinity");
467 }
468 else
469 {
470 md_print_warn(cr, fplog,
471 "Non-default thread affinity set probably by the OpenMP library,\n"
472 "disabling internal thread affinity");
473 }
474 hw_opt->thread_affinity = threadaffOFF;
475 }
476 else
477 {
478 /* Only warn once, at the last check (bAfterOpenmpInit==TRUE) */
479 if (bAfterOpenmpInit)
480 {
481 md_print_warn(cr, fplog,
482 "Overriding thread affinity set outside %s\n",
483 ShortProgram());
484 }
485 }
486
487 if (debug)
488 {
489 fprintf(debug, "Non-default affinity mask found\n");
490 }
491 }
492 else
493 {
494 if (debug)
495 {
496 fprintf(debug, "Default affinity mask found\n");
497 }
498 }
499#endif /* HAVE_SCHED_GETAFFINITY */
500}