summaryrefslogtreecommitdiff
path: root/libgomp/loop.c
blob: ca389214c2f6debb06a6886e7c4bca4b34e2e6e7 (plain)
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
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
/* Copyright (C) 2005, 2008, 2009 Free Software Foundation, Inc.
   Contributed by Richard Henderson <rth@redhat.com>.

   This file is part of the GNU OpenMP Library (libgomp).

   Libgomp is free software; you can redistribute it and/or modify it
   under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 3, or (at your option)
   any later version.

   Libgomp is distributed in the hope that it will be useful, but WITHOUT ANY
   WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
   FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
   more details.

   Under Section 7 of GPL version 3, you are granted additional
   permissions described in the GCC Runtime Library Exception, version
   3.1, as published by the Free Software Foundation.

   You should have received a copy of the GNU General Public License and
   a copy of the GCC Runtime Library Exception along with this program;
   see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
   <http://www.gnu.org/licenses/>.  */

/* This file handles the LOOP (FOR/DO) construct.  */

#include <limits.h>
#include <stdlib.h>
#include "libgomp.h"


/* Initialize the given work share construct from the given arguments.  */

static inline void
gomp_loop_init (struct gomp_work_share *ws, long start, long end, long incr,
		enum gomp_schedule_type sched, long chunk_size)
{
  ws->sched = sched;
  ws->chunk_size = chunk_size;
  /* Canonicalize loops that have zero iterations to ->next == ->end.  */
  ws->end = ((incr > 0 && start > end) || (incr < 0 && start < end))
	    ? start : end;
  ws->incr = incr;
  ws->next = start;
  if (sched == GFS_DYNAMIC)
    {
      ws->chunk_size *= incr;

#ifdef HAVE_SYNC_BUILTINS
      {
	/* For dynamic scheduling prepare things to make each iteration
	   faster.  */
	struct gomp_thread *thr = gomp_thread ();
	struct gomp_team *team = thr->ts.team;
	long nthreads = team ? team->nthreads : 1;

	if (__builtin_expect (incr > 0, 1))
	  {
	    /* Cheap overflow protection.  */
	    if (__builtin_expect ((nthreads | ws->chunk_size)
				  >= 1UL << (sizeof (long)
					     * __CHAR_BIT__ / 2 - 1), 0))
	      ws->mode = 0;
	    else
	      ws->mode = ws->end < (LONG_MAX
				    - (nthreads + 1) * ws->chunk_size);
	  }
	/* Cheap overflow protection.  */
	else if (__builtin_expect ((nthreads | -ws->chunk_size)
				   >= 1UL << (sizeof (long)
					      * __CHAR_BIT__ / 2 - 1), 0))
	  ws->mode = 0;
	else
	  ws->mode = ws->end > (nthreads + 1) * -ws->chunk_size - LONG_MAX;
      }
#endif
    }
}

/* The *_start routines are called when first encountering a loop construct
   that is not bound directly to a parallel construct.  The first thread 
   that arrives will create the work-share construct; subsequent threads
   will see the construct exists and allocate work from it.

   START, END, INCR are the bounds of the loop; due to the restrictions of
   OpenMP, these values must be the same in every thread.  This is not 
   verified (nor is it entirely verifiable, since START is not necessarily
   retained intact in the work-share data structure).  CHUNK_SIZE is the
   scheduling parameter; again this must be identical in all threads.

   Returns true if there's any work for this thread to perform.  If so,
   *ISTART and *IEND are filled with the bounds of the iteration block
   allocated to this thread.  Returns false if all work was assigned to
   other threads prior to this thread's arrival.  */

static bool
gomp_loop_static_start (long start, long end, long incr, long chunk_size,
			long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();

  thr->ts.static_trip = 0;
  if (gomp_work_share_start (false))
    {
      gomp_loop_init (thr->ts.work_share, start, end, incr,
		      GFS_STATIC, chunk_size);
      gomp_work_share_init_done ();
    }

  return !gomp_iter_static_next (istart, iend);
}

static bool
gomp_loop_dynamic_start (long start, long end, long incr, long chunk_size,
			 long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  bool ret;

  if (gomp_work_share_start (false))
    {
      gomp_loop_init (thr->ts.work_share, start, end, incr,
		      GFS_DYNAMIC, chunk_size);
      gomp_work_share_init_done ();
    }

#ifdef HAVE_SYNC_BUILTINS
  ret = gomp_iter_dynamic_next (istart, iend);
#else
  gomp_mutex_lock (&thr->ts.work_share->lock);
  ret = gomp_iter_dynamic_next_locked (istart, iend);
  gomp_mutex_unlock (&thr->ts.work_share->lock);
#endif

  return ret;
}

static bool
gomp_loop_guided_start (long start, long end, long incr, long chunk_size,
			long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  bool ret;

  if (gomp_work_share_start (false))
    {
      gomp_loop_init (thr->ts.work_share, start, end, incr,
		      GFS_GUIDED, chunk_size);
      gomp_work_share_init_done ();
    }

#ifdef HAVE_SYNC_BUILTINS
  ret = gomp_iter_guided_next (istart, iend);
#else
  gomp_mutex_lock (&thr->ts.work_share->lock);
  ret = gomp_iter_guided_next_locked (istart, iend);
  gomp_mutex_unlock (&thr->ts.work_share->lock);
#endif

  return ret;
}

bool
GOMP_loop_runtime_start (long start, long end, long incr,
			 long *istart, long *iend)
{
  struct gomp_task_icv *icv = gomp_icv (false);
  switch (icv->run_sched_var)
    {
    case GFS_STATIC:
      return gomp_loop_static_start (start, end, incr, icv->run_sched_modifier,
				     istart, iend);
    case GFS_DYNAMIC:
      return gomp_loop_dynamic_start (start, end, incr, icv->run_sched_modifier,
				      istart, iend);
    case GFS_GUIDED:
      return gomp_loop_guided_start (start, end, incr, icv->run_sched_modifier,
				     istart, iend);
    case GFS_AUTO:
      /* For now map to schedule(static), later on we could play with feedback
	 driven choice.  */
      return gomp_loop_static_start (start, end, incr, 0, istart, iend);
    default:
      abort ();
    }
}

/* The *_ordered_*_start routines are similar.  The only difference is that
   this work-share construct is initialized to expect an ORDERED section.  */

static bool
gomp_loop_ordered_static_start (long start, long end, long incr,
				long chunk_size, long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();

  thr->ts.static_trip = 0;
  if (gomp_work_share_start (true))
    {
      gomp_loop_init (thr->ts.work_share, start, end, incr,
		      GFS_STATIC, chunk_size);
      gomp_ordered_static_init ();
      gomp_work_share_init_done ();
    }

  return !gomp_iter_static_next (istart, iend);
}

static bool
gomp_loop_ordered_dynamic_start (long start, long end, long incr,
				 long chunk_size, long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  bool ret;

  if (gomp_work_share_start (true))
    {
      gomp_loop_init (thr->ts.work_share, start, end, incr,
		      GFS_DYNAMIC, chunk_size);
      gomp_mutex_lock (&thr->ts.work_share->lock);
      gomp_work_share_init_done ();
    }
  else
    gomp_mutex_lock (&thr->ts.work_share->lock);

  ret = gomp_iter_dynamic_next_locked (istart, iend);
  if (ret)
    gomp_ordered_first ();
  gomp_mutex_unlock (&thr->ts.work_share->lock);

  return ret;
}

static bool
gomp_loop_ordered_guided_start (long start, long end, long incr,
				long chunk_size, long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  bool ret;

  if (gomp_work_share_start (true))
    {
      gomp_loop_init (thr->ts.work_share, start, end, incr,
		      GFS_GUIDED, chunk_size);
      gomp_mutex_lock (&thr->ts.work_share->lock);
      gomp_work_share_init_done ();
    }
  else
    gomp_mutex_lock (&thr->ts.work_share->lock);

  ret = gomp_iter_guided_next_locked (istart, iend);
  if (ret)
    gomp_ordered_first ();
  gomp_mutex_unlock (&thr->ts.work_share->lock);

  return ret;
}

bool
GOMP_loop_ordered_runtime_start (long start, long end, long incr,
				 long *istart, long *iend)
{
  struct gomp_task_icv *icv = gomp_icv (false);
  switch (icv->run_sched_var)
    {
    case GFS_STATIC:
      return gomp_loop_ordered_static_start (start, end, incr,
					     icv->run_sched_modifier,
					     istart, iend);
    case GFS_DYNAMIC:
      return gomp_loop_ordered_dynamic_start (start, end, incr,
					      icv->run_sched_modifier,
					      istart, iend);
    case GFS_GUIDED:
      return gomp_loop_ordered_guided_start (start, end, incr,
					     icv->run_sched_modifier,
					     istart, iend);
    case GFS_AUTO:
      /* For now map to schedule(static), later on we could play with feedback
	 driven choice.  */
      return gomp_loop_ordered_static_start (start, end, incr,
					     0, istart, iend);
    default:
      abort ();
    }
}

/* The *_next routines are called when the thread completes processing of 
   the iteration block currently assigned to it.  If the work-share 
   construct is bound directly to a parallel construct, then the iteration
   bounds may have been set up before the parallel.  In which case, this
   may be the first iteration for the thread.

   Returns true if there is work remaining to be performed; *ISTART and
   *IEND are filled with a new iteration block.  Returns false if all work
   has been assigned.  */

static bool
gomp_loop_static_next (long *istart, long *iend)
{
  return !gomp_iter_static_next (istart, iend);
}

static bool
gomp_loop_dynamic_next (long *istart, long *iend)
{
  bool ret;

#ifdef HAVE_SYNC_BUILTINS
  ret = gomp_iter_dynamic_next (istart, iend);
#else
  struct gomp_thread *thr = gomp_thread ();
  gomp_mutex_lock (&thr->ts.work_share->lock);
  ret = gomp_iter_dynamic_next_locked (istart, iend);
  gomp_mutex_unlock (&thr->ts.work_share->lock);
#endif

  return ret;
}

static bool
gomp_loop_guided_next (long *istart, long *iend)
{
  bool ret;

#ifdef HAVE_SYNC_BUILTINS
  ret = gomp_iter_guided_next (istart, iend);
#else
  struct gomp_thread *thr = gomp_thread ();
  gomp_mutex_lock (&thr->ts.work_share->lock);
  ret = gomp_iter_guided_next_locked (istart, iend);
  gomp_mutex_unlock (&thr->ts.work_share->lock);
#endif

  return ret;
}

bool
GOMP_loop_runtime_next (long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  
  switch (thr->ts.work_share->sched)
    {
    case GFS_STATIC:
    case GFS_AUTO:
      return gomp_loop_static_next (istart, iend);
    case GFS_DYNAMIC:
      return gomp_loop_dynamic_next (istart, iend);
    case GFS_GUIDED:
      return gomp_loop_guided_next (istart, iend);
    default:
      abort ();
    }
}

/* The *_ordered_*_next routines are called when the thread completes
   processing of the iteration block currently assigned to it.

   Returns true if there is work remaining to be performed; *ISTART and
   *IEND are filled with a new iteration block.  Returns false if all work
   has been assigned.  */

static bool
gomp_loop_ordered_static_next (long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  int test;

  gomp_ordered_sync ();
  gomp_mutex_lock (&thr->ts.work_share->lock);
  test = gomp_iter_static_next (istart, iend);
  if (test >= 0)
    gomp_ordered_static_next ();
  gomp_mutex_unlock (&thr->ts.work_share->lock);

  return test == 0;
}

static bool
gomp_loop_ordered_dynamic_next (long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  bool ret;

  gomp_ordered_sync ();
  gomp_mutex_lock (&thr->ts.work_share->lock);
  ret = gomp_iter_dynamic_next_locked (istart, iend);
  if (ret)
    gomp_ordered_next ();
  else
    gomp_ordered_last ();
  gomp_mutex_unlock (&thr->ts.work_share->lock);

  return ret;
}

static bool
gomp_loop_ordered_guided_next (long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  bool ret;

  gomp_ordered_sync ();
  gomp_mutex_lock (&thr->ts.work_share->lock);
  ret = gomp_iter_guided_next_locked (istart, iend);
  if (ret)
    gomp_ordered_next ();
  else
    gomp_ordered_last ();
  gomp_mutex_unlock (&thr->ts.work_share->lock);

  return ret;
}

bool
GOMP_loop_ordered_runtime_next (long *istart, long *iend)
{
  struct gomp_thread *thr = gomp_thread ();
  
  switch (thr->ts.work_share->sched)
    {
    case GFS_STATIC:
    case GFS_AUTO:
      return gomp_loop_ordered_static_next (istart, iend);
    case GFS_DYNAMIC:
      return gomp_loop_ordered_dynamic_next (istart, iend);
    case GFS_GUIDED:
      return gomp_loop_ordered_guided_next (istart, iend);
    default:
      abort ();
    }
}

/* The GOMP_parallel_loop_* routines pre-initialize a work-share construct
   to avoid one synchronization once we get into the loop.  */

static void
gomp_parallel_loop_start (void (*fn) (void *), void *data,
			  unsigned num_threads, long start, long end,
			  long incr, enum gomp_schedule_type sched,
			  long chunk_size)
{
  struct gomp_team *team;

  num_threads = gomp_resolve_num_threads (num_threads, 0);
  team = gomp_new_team (num_threads);
  gomp_loop_init (&team->work_shares[0], start, end, incr, sched, chunk_size);
  gomp_team_start (fn, data, num_threads, team);
}

void
GOMP_parallel_loop_static_start (void (*fn) (void *), void *data,
				 unsigned num_threads, long start, long end,
				 long incr, long chunk_size)
{
  gomp_parallel_loop_start (fn, data, num_threads, start, end, incr,
			    GFS_STATIC, chunk_size);
}

void
GOMP_parallel_loop_dynamic_start (void (*fn) (void *), void *data,
				  unsigned num_threads, long start, long end,
				  long incr, long chunk_size)
{
  gomp_parallel_loop_start (fn, data, num_threads, start, end, incr,
			    GFS_DYNAMIC, chunk_size);
}

void
GOMP_parallel_loop_guided_start (void (*fn) (void *), void *data,
				 unsigned num_threads, long start, long end,
				 long incr, long chunk_size)
{
  gomp_parallel_loop_start (fn, data, num_threads, start, end, incr,
			    GFS_GUIDED, chunk_size);
}

void
GOMP_parallel_loop_runtime_start (void (*fn) (void *), void *data,
				  unsigned num_threads, long start, long end,
				  long incr)
{
  struct gomp_task_icv *icv = gomp_icv (false);
  gomp_parallel_loop_start (fn, data, num_threads, start, end, incr,
			    icv->run_sched_var, icv->run_sched_modifier);
}

/* The GOMP_loop_end* routines are called after the thread is told that
   all loop iterations are complete.  This first version synchronizes
   all threads; the nowait version does not.  */

void
GOMP_loop_end (void)
{
  gomp_work_share_end ();
}

void
GOMP_loop_end_nowait (void)
{
  gomp_work_share_end_nowait ();
}


/* We use static functions above so that we're sure that the "runtime"
   function can defer to the proper routine without interposition.  We
   export the static function with a strong alias when possible, or with
   a wrapper function otherwise.  */

#ifdef HAVE_ATTRIBUTE_ALIAS
extern __typeof(gomp_loop_static_start) GOMP_loop_static_start
	__attribute__((alias ("gomp_loop_static_start")));
extern __typeof(gomp_loop_dynamic_start) GOMP_loop_dynamic_start
	__attribute__((alias ("gomp_loop_dynamic_start")));
extern __typeof(gomp_loop_guided_start) GOMP_loop_guided_start
	__attribute__((alias ("gomp_loop_guided_start")));

extern __typeof(gomp_loop_ordered_static_start) GOMP_loop_ordered_static_start
	__attribute__((alias ("gomp_loop_ordered_static_start")));
extern __typeof(gomp_loop_ordered_dynamic_start) GOMP_loop_ordered_dynamic_start
	__attribute__((alias ("gomp_loop_ordered_dynamic_start")));
extern __typeof(gomp_loop_ordered_guided_start) GOMP_loop_ordered_guided_start
	__attribute__((alias ("gomp_loop_ordered_guided_start")));

extern __typeof(gomp_loop_static_next) GOMP_loop_static_next
	__attribute__((alias ("gomp_loop_static_next")));
extern __typeof(gomp_loop_dynamic_next) GOMP_loop_dynamic_next
	__attribute__((alias ("gomp_loop_dynamic_next")));
extern __typeof(gomp_loop_guided_next) GOMP_loop_guided_next
	__attribute__((alias ("gomp_loop_guided_next")));

extern __typeof(gomp_loop_ordered_static_next) GOMP_loop_ordered_static_next
	__attribute__((alias ("gomp_loop_ordered_static_next")));
extern __typeof(gomp_loop_ordered_dynamic_next) GOMP_loop_ordered_dynamic_next
	__attribute__((alias ("gomp_loop_ordered_dynamic_next")));
extern __typeof(gomp_loop_ordered_guided_next) GOMP_loop_ordered_guided_next
	__attribute__((alias ("gomp_loop_ordered_guided_next")));
#else
bool
GOMP_loop_static_start (long start, long end, long incr, long chunk_size,
			long *istart, long *iend)
{
  return gomp_loop_static_start (start, end, incr, chunk_size, istart, iend);
}

bool
GOMP_loop_dynamic_start (long start, long end, long incr, long chunk_size,
			 long *istart, long *iend)
{
  return gomp_loop_dynamic_start (start, end, incr, chunk_size, istart, iend);
}

bool
GOMP_loop_guided_start (long start, long end, long incr, long chunk_size,
			long *istart, long *iend)
{
  return gomp_loop_guided_start (start, end, incr, chunk_size, istart, iend);
}

bool
GOMP_loop_ordered_static_start (long start, long end, long incr,
				long chunk_size, long *istart, long *iend)
{
  return gomp_loop_ordered_static_start (start, end, incr, chunk_size,
					 istart, iend);
}

bool
GOMP_loop_ordered_dynamic_start (long start, long end, long incr,
				 long chunk_size, long *istart, long *iend)
{
  return gomp_loop_ordered_dynamic_start (start, end, incr, chunk_size,
					  istart, iend);
}

bool
GOMP_loop_ordered_guided_start (long start, long end, long incr,
				long chunk_size, long *istart, long *iend)
{
  return gomp_loop_ordered_guided_start (start, end, incr, chunk_size,
					 istart, iend);
}

bool
GOMP_loop_static_next (long *istart, long *iend)
{
  return gomp_loop_static_next (istart, iend);
}

bool
GOMP_loop_dynamic_next (long *istart, long *iend)
{
  return gomp_loop_dynamic_next (istart, iend);
}

bool
GOMP_loop_guided_next (long *istart, long *iend)
{
  return gomp_loop_guided_next (istart, iend);
}

bool
GOMP_loop_ordered_static_next (long *istart, long *iend)
{
  return gomp_loop_ordered_static_next (istart, iend);
}

bool
GOMP_loop_ordered_dynamic_next (long *istart, long *iend)
{
  return gomp_loop_ordered_dynamic_next (istart, iend);
}

bool
GOMP_loop_ordered_guided_next (long *istart, long *iend)
{
  return gomp_loop_ordered_guided_next (istart, iend);
}
#endif