summaryrefslogtreecommitdiff
path: root/gcc/tree-vectorizer.c
blob: 584f13095ab11fb6568b7e18d30a11b53ae4e052 (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
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
/* Vectorizer
   Copyright (C) 2003-2020 Free Software Foundation, Inc.
   Contributed by Dorit Naishlos <dorit@il.ibm.com>

This file is part of GCC.

GCC 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.

GCC 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.

You should have received a copy of the GNU General Public License
along with GCC; see the file COPYING3.  If not see
<http://www.gnu.org/licenses/>.  */

/* Loop and basic block vectorizer.

  This file contains drivers for the three vectorizers:
  (1) loop vectorizer (inter-iteration parallelism),
  (2) loop-aware SLP (intra-iteration parallelism) (invoked by the loop
      vectorizer)
  (3) BB vectorizer (out-of-loops), aka SLP

  The rest of the vectorizer's code is organized as follows:
  - tree-vect-loop.c - loop specific parts such as reductions, etc. These are
    used by drivers (1) and (2).
  - tree-vect-loop-manip.c - vectorizer's loop control-flow utilities, used by
    drivers (1) and (2).
  - tree-vect-slp.c - BB vectorization specific analysis and transformation,
    used by drivers (2) and (3).
  - tree-vect-stmts.c - statements analysis and transformation (used by all).
  - tree-vect-data-refs.c - vectorizer specific data-refs analysis and
    manipulations (used by all).
  - tree-vect-patterns.c - vectorizable code patterns detector (used by all)

  Here's a poor attempt at illustrating that:

     tree-vectorizer.c:
     loop_vect()  loop_aware_slp()  slp_vect()
          |        /           \          /
          |       /             \        /
          tree-vect-loop.c  tree-vect-slp.c
                | \      \  /      /   |
                |  \      \/      /    |
                |   \     /\     /     |
                |    \   /  \   /      |
         tree-vect-stmts.c  tree-vect-data-refs.c
                       \      /
                    tree-vect-patterns.c
*/

#include "config.h"
#include "system.h"
#include "coretypes.h"
#include "backend.h"
#include "tree.h"
#include "gimple.h"
#include "predict.h"
#include "tree-pass.h"
#include "ssa.h"
#include "cgraph.h"
#include "fold-const.h"
#include "stor-layout.h"
#include "gimple-iterator.h"
#include "gimple-walk.h"
#include "tree-ssa-loop-manip.h"
#include "tree-ssa-loop-niter.h"
#include "tree-cfg.h"
#include "cfgloop.h"
#include "tree-vectorizer.h"
#include "tree-ssa-propagate.h"
#include "dbgcnt.h"
#include "tree-scalar-evolution.h"
#include "stringpool.h"
#include "attribs.h"
#include "gimple-pretty-print.h"
#include "opt-problem.h"
#include "internal-fn.h"


/* Loop or bb location, with hotness information.  */
dump_user_location_t vect_location;

/* auto_purge_vect_location's dtor: reset the vect_location
   global, to avoid stale location_t values that could reference
   GC-ed blocks.  */

auto_purge_vect_location::~auto_purge_vect_location ()
{
  vect_location = dump_user_location_t ();
}

/* Dump a cost entry according to args to F.  */

void
dump_stmt_cost (FILE *f, void *data, int count, enum vect_cost_for_stmt kind,
		stmt_vec_info stmt_info, tree, int misalign, unsigned cost,
		enum vect_cost_model_location where)
{
  fprintf (f, "%p ", data);
  if (stmt_info)
    {
      print_gimple_expr (f, STMT_VINFO_STMT (stmt_info), 0, TDF_SLIM);
      fprintf (f, " ");
    }
  else
    fprintf (f, "<unknown> ");
  fprintf (f, "%d times ", count);
  const char *ks = "unknown";
  switch (kind)
    {
    case scalar_stmt:
      ks = "scalar_stmt";
      break;
    case scalar_load:
      ks = "scalar_load";
      break;
    case scalar_store:
      ks = "scalar_store";
      break;
    case vector_stmt:
      ks = "vector_stmt";
      break;
    case vector_load:
      ks = "vector_load";
      break;
    case vector_gather_load:
      ks = "vector_gather_load";
      break;
    case unaligned_load:
      ks = "unaligned_load";
      break;
    case unaligned_store:
      ks = "unaligned_store";
      break;
    case vector_store:
      ks = "vector_store";
      break;
    case vector_scatter_store:
      ks = "vector_scatter_store";
      break;
    case vec_to_scalar:
      ks = "vec_to_scalar";
      break;
    case scalar_to_vec:
      ks = "scalar_to_vec";
      break;
    case cond_branch_not_taken:
      ks = "cond_branch_not_taken";
      break;
    case cond_branch_taken:
      ks = "cond_branch_taken";
      break;
    case vec_perm:
      ks = "vec_perm";
      break;
    case vec_promote_demote:
      ks = "vec_promote_demote";
      break;
    case vec_construct:
      ks = "vec_construct";
      break;
    }
  fprintf (f, "%s ", ks);
  if (kind == unaligned_load || kind == unaligned_store)
    fprintf (f, "(misalign %d) ", misalign);
  fprintf (f, "costs %u ", cost);
  const char *ws = "unknown";
  switch (where)
    {
    case vect_prologue:
      ws = "prologue";
      break;
    case vect_body:
      ws = "body";
      break;
    case vect_epilogue:
      ws = "epilogue";
      break;
    }
  fprintf (f, "in %s\n", ws);
}

/* For mapping simduid to vectorization factor.  */

class simduid_to_vf : public free_ptr_hash<simduid_to_vf>
{
public:
  unsigned int simduid;
  poly_uint64 vf;

  /* hash_table support.  */
  static inline hashval_t hash (const simduid_to_vf *);
  static inline int equal (const simduid_to_vf *, const simduid_to_vf *);
};

inline hashval_t
simduid_to_vf::hash (const simduid_to_vf *p)
{
  return p->simduid;
}

inline int
simduid_to_vf::equal (const simduid_to_vf *p1, const simduid_to_vf *p2)
{
  return p1->simduid == p2->simduid;
}

/* This hash maps the OMP simd array to the corresponding simduid used
   to index into it.  Like thus,

        _7 = GOMP_SIMD_LANE (simduid.0)
        ...
        ...
        D.1737[_7] = stuff;


   This hash maps from the OMP simd array (D.1737[]) to DECL_UID of
   simduid.0.  */

struct simd_array_to_simduid : free_ptr_hash<simd_array_to_simduid>
{
  tree decl;
  unsigned int simduid;

  /* hash_table support.  */
  static inline hashval_t hash (const simd_array_to_simduid *);
  static inline int equal (const simd_array_to_simduid *,
			   const simd_array_to_simduid *);
};

inline hashval_t
simd_array_to_simduid::hash (const simd_array_to_simduid *p)
{
  return DECL_UID (p->decl);
}

inline int
simd_array_to_simduid::equal (const simd_array_to_simduid *p1,
			      const simd_array_to_simduid *p2)
{
  return p1->decl == p2->decl;
}

/* Fold IFN_GOMP_SIMD_LANE, IFN_GOMP_SIMD_VF, IFN_GOMP_SIMD_LAST_LANE,
   into their corresponding constants and remove
   IFN_GOMP_SIMD_ORDERED_{START,END}.  */

static void
adjust_simduid_builtins (hash_table<simduid_to_vf> *htab)
{
  basic_block bb;

  FOR_EACH_BB_FN (bb, cfun)
    {
      gimple_stmt_iterator i;

      for (i = gsi_start_bb (bb); !gsi_end_p (i); )
	{
	  poly_uint64 vf = 1;
	  enum internal_fn ifn;
	  gimple *stmt = gsi_stmt (i);
	  tree t;
	  if (!is_gimple_call (stmt)
	      || !gimple_call_internal_p (stmt))
	    {
	      gsi_next (&i);
	      continue;
	    }
	  ifn = gimple_call_internal_fn (stmt);
	  switch (ifn)
	    {
	    case IFN_GOMP_SIMD_LANE:
	    case IFN_GOMP_SIMD_VF:
	    case IFN_GOMP_SIMD_LAST_LANE:
	      break;
	    case IFN_GOMP_SIMD_ORDERED_START:
	    case IFN_GOMP_SIMD_ORDERED_END:
	      if (integer_onep (gimple_call_arg (stmt, 0)))
		{
		  enum built_in_function bcode
		    = (ifn == IFN_GOMP_SIMD_ORDERED_START
		       ? BUILT_IN_GOMP_ORDERED_START
		       : BUILT_IN_GOMP_ORDERED_END);
		  gimple *g
		    = gimple_build_call (builtin_decl_explicit (bcode), 0);
		  gimple_move_vops (g, stmt);
		  gsi_replace (&i, g, true);
		  continue;
		}
	      gsi_remove (&i, true);
	      unlink_stmt_vdef (stmt);
	      continue;
	    default:
	      gsi_next (&i);
	      continue;
	    }
	  tree arg = gimple_call_arg (stmt, 0);
	  gcc_assert (arg != NULL_TREE);
	  gcc_assert (TREE_CODE (arg) == SSA_NAME);
	  simduid_to_vf *p = NULL, data;
	  data.simduid = DECL_UID (SSA_NAME_VAR (arg));
	  /* Need to nullify loop safelen field since it's value is not
	     valid after transformation.  */
	  if (bb->loop_father && bb->loop_father->safelen > 0)
	    bb->loop_father->safelen = 0;
	  if (htab)
	    {
	      p = htab->find (&data);
	      if (p)
		vf = p->vf;
	    }
	  switch (ifn)
	    {
	    case IFN_GOMP_SIMD_VF:
	      t = build_int_cst (unsigned_type_node, vf);
	      break;
	    case IFN_GOMP_SIMD_LANE:
	      t = build_int_cst (unsigned_type_node, 0);
	      break;
	    case IFN_GOMP_SIMD_LAST_LANE:
	      t = gimple_call_arg (stmt, 1);
	      break;
	    default:
	      gcc_unreachable ();
	    }
	  tree lhs = gimple_call_lhs (stmt);
	  if (lhs)
	    replace_uses_by (lhs, t);
	  release_defs (stmt);
	  gsi_remove (&i, true);
	}
    }
}

/* Helper structure for note_simd_array_uses.  */

struct note_simd_array_uses_struct
{
  hash_table<simd_array_to_simduid> **htab;
  unsigned int simduid;
};

/* Callback for note_simd_array_uses, called through walk_gimple_op.  */

static tree
note_simd_array_uses_cb (tree *tp, int *walk_subtrees, void *data)
{
  struct walk_stmt_info *wi = (struct walk_stmt_info *) data;
  struct note_simd_array_uses_struct *ns
    = (struct note_simd_array_uses_struct *) wi->info;

  if (TYPE_P (*tp))
    *walk_subtrees = 0;
  else if (VAR_P (*tp)
	   && lookup_attribute ("omp simd array", DECL_ATTRIBUTES (*tp))
	   && DECL_CONTEXT (*tp) == current_function_decl)
    {
      simd_array_to_simduid data;
      if (!*ns->htab)
	*ns->htab = new hash_table<simd_array_to_simduid> (15);
      data.decl = *tp;
      data.simduid = ns->simduid;
      simd_array_to_simduid **slot = (*ns->htab)->find_slot (&data, INSERT);
      if (*slot == NULL)
	{
	  simd_array_to_simduid *p = XNEW (simd_array_to_simduid);
	  *p = data;
	  *slot = p;
	}
      else if ((*slot)->simduid != ns->simduid)
	(*slot)->simduid = -1U;
      *walk_subtrees = 0;
    }
  return NULL_TREE;
}

/* Find "omp simd array" temporaries and map them to corresponding
   simduid.  */

static void
note_simd_array_uses (hash_table<simd_array_to_simduid> **htab)
{
  basic_block bb;
  gimple_stmt_iterator gsi;
  struct walk_stmt_info wi;
  struct note_simd_array_uses_struct ns;

  memset (&wi, 0, sizeof (wi));
  wi.info = &ns;
  ns.htab = htab;

  FOR_EACH_BB_FN (bb, cfun)
    for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi))
      {
	gimple *stmt = gsi_stmt (gsi);
	if (!is_gimple_call (stmt) || !gimple_call_internal_p (stmt))
	  continue;
	switch (gimple_call_internal_fn (stmt))
	  {
	  case IFN_GOMP_SIMD_LANE:
	  case IFN_GOMP_SIMD_VF:
	  case IFN_GOMP_SIMD_LAST_LANE:
	    break;
	  default:
	    continue;
	  }
	tree lhs = gimple_call_lhs (stmt);
	if (lhs == NULL_TREE)
	  continue;
	imm_use_iterator use_iter;
	gimple *use_stmt;
	ns.simduid = DECL_UID (SSA_NAME_VAR (gimple_call_arg (stmt, 0)));
	FOR_EACH_IMM_USE_STMT (use_stmt, use_iter, lhs)
	  if (!is_gimple_debug (use_stmt))
	    walk_gimple_op (use_stmt, note_simd_array_uses_cb, &wi);
      }
}

/* Shrink arrays with "omp simd array" attribute to the corresponding
   vectorization factor.  */

static void
shrink_simd_arrays
  (hash_table<simd_array_to_simduid> *simd_array_to_simduid_htab,
   hash_table<simduid_to_vf> *simduid_to_vf_htab)
{
  for (hash_table<simd_array_to_simduid>::iterator iter
	 = simd_array_to_simduid_htab->begin ();
       iter != simd_array_to_simduid_htab->end (); ++iter)
    if ((*iter)->simduid != -1U)
      {
	tree decl = (*iter)->decl;
	poly_uint64 vf = 1;
	if (simduid_to_vf_htab)
	  {
	    simduid_to_vf *p = NULL, data;
	    data.simduid = (*iter)->simduid;
	    p = simduid_to_vf_htab->find (&data);
	    if (p)
	      vf = p->vf;
	  }
	tree atype
	  = build_array_type_nelts (TREE_TYPE (TREE_TYPE (decl)), vf);
	TREE_TYPE (decl) = atype;
	relayout_decl (decl);
      }

  delete simd_array_to_simduid_htab;
}

/* Initialize the vec_info with kind KIND_IN and target cost data
   TARGET_COST_DATA_IN.  */

vec_info::vec_info (vec_info::vec_kind kind_in, void *target_cost_data_in,
		    vec_info_shared *shared_)
  : kind (kind_in),
    shared (shared_),
    target_cost_data (target_cost_data_in)
{
  stmt_vec_infos.create (50);
}

vec_info::~vec_info ()
{
  slp_instance instance;
  unsigned int i;

  FOR_EACH_VEC_ELT (slp_instances, i, instance)
    vect_free_slp_instance (instance, true);

  destroy_cost_data (target_cost_data);
  free_stmt_vec_infos ();
}

vec_info_shared::vec_info_shared ()
  : datarefs (vNULL),
    datarefs_copy (vNULL),
    ddrs (vNULL)
{
}

vec_info_shared::~vec_info_shared ()
{
  free_data_refs (datarefs);
  free_dependence_relations (ddrs);
  datarefs_copy.release ();
}

void
vec_info_shared::save_datarefs ()
{
  if (!flag_checking)
    return;
  datarefs_copy.reserve_exact (datarefs.length ());
  for (unsigned i = 0; i < datarefs.length (); ++i)
    datarefs_copy.quick_push (*datarefs[i]);
}

void
vec_info_shared::check_datarefs ()
{
  if (!flag_checking)
    return;
  gcc_assert (datarefs.length () == datarefs_copy.length ());
  for (unsigned i = 0; i < datarefs.length (); ++i)
    if (memcmp (&datarefs_copy[i], datarefs[i], sizeof (data_reference)) != 0)
      gcc_unreachable ();
}

/* Record that STMT belongs to the vectorizable region.  Create and return
   an associated stmt_vec_info.  */

stmt_vec_info
vec_info::add_stmt (gimple *stmt)
{
  stmt_vec_info res = new_stmt_vec_info (stmt);
  set_vinfo_for_stmt (stmt, res);
  return res;
}

/* If STMT has an associated stmt_vec_info, return that vec_info, otherwise
   return null.  It is safe to call this function on any statement, even if
   it might not be part of the vectorizable region.  */

stmt_vec_info
vec_info::lookup_stmt (gimple *stmt)
{
  unsigned int uid = gimple_uid (stmt);
  if (uid > 0 && uid - 1 < stmt_vec_infos.length ())
    {
      stmt_vec_info res = stmt_vec_infos[uid - 1];
      if (res && res->stmt == stmt)
	return res;
    }
  return NULL;
}

/* If NAME is an SSA_NAME and its definition has an associated stmt_vec_info,
   return that stmt_vec_info, otherwise return null.  It is safe to call
   this on arbitrary operands.  */

stmt_vec_info
vec_info::lookup_def (tree name)
{
  if (TREE_CODE (name) == SSA_NAME
      && !SSA_NAME_IS_DEFAULT_DEF (name))
    return lookup_stmt (SSA_NAME_DEF_STMT (name));
  return NULL;
}

/* See whether there is a single non-debug statement that uses LHS and
   whether that statement has an associated stmt_vec_info.  Return the
   stmt_vec_info if so, otherwise return null.  */

stmt_vec_info
vec_info::lookup_single_use (tree lhs)
{
  use_operand_p dummy;
  gimple *use_stmt;
  if (single_imm_use (lhs, &dummy, &use_stmt))
    return lookup_stmt (use_stmt);
  return NULL;
}

/* Return vectorization information about DR.  */

dr_vec_info *
vec_info::lookup_dr (data_reference *dr)
{
  stmt_vec_info stmt_info = lookup_stmt (DR_STMT (dr));
  /* DR_STMT should never refer to a stmt in a pattern replacement.  */
  gcc_checking_assert (!is_pattern_stmt_p (stmt_info));
  return STMT_VINFO_DR_INFO (stmt_info->dr_aux.stmt);
}

/* Record that NEW_STMT_INFO now implements the same data reference
   as OLD_STMT_INFO.  */

void
vec_info::move_dr (stmt_vec_info new_stmt_info, stmt_vec_info old_stmt_info)
{
  gcc_assert (!is_pattern_stmt_p (old_stmt_info));
  STMT_VINFO_DR_INFO (old_stmt_info)->stmt = new_stmt_info;
  new_stmt_info->dr_aux = old_stmt_info->dr_aux;
  STMT_VINFO_DR_WRT_VEC_LOOP (new_stmt_info)
    = STMT_VINFO_DR_WRT_VEC_LOOP (old_stmt_info);
  STMT_VINFO_GATHER_SCATTER_P (new_stmt_info)
    = STMT_VINFO_GATHER_SCATTER_P (old_stmt_info);
}

/* Permanently remove the statement described by STMT_INFO from the
   function.  */

void
vec_info::remove_stmt (stmt_vec_info stmt_info)
{
  gcc_assert (!stmt_info->pattern_stmt_p);
  set_vinfo_for_stmt (stmt_info->stmt, NULL);
  gimple_stmt_iterator si = gsi_for_stmt (stmt_info->stmt);
  unlink_stmt_vdef (stmt_info->stmt);
  gsi_remove (&si, true);
  release_defs (stmt_info->stmt);
  free_stmt_vec_info (stmt_info);
}

/* Replace the statement at GSI by NEW_STMT, both the vectorization
   information and the function itself.  STMT_INFO describes the statement
   at GSI.  */

void
vec_info::replace_stmt (gimple_stmt_iterator *gsi, stmt_vec_info stmt_info,
			gimple *new_stmt)
{
  gimple *old_stmt = stmt_info->stmt;
  gcc_assert (!stmt_info->pattern_stmt_p && old_stmt == gsi_stmt (*gsi));
  set_vinfo_for_stmt (old_stmt, NULL);
  set_vinfo_for_stmt (new_stmt, stmt_info);
  stmt_info->stmt = new_stmt;
  gsi_replace (gsi, new_stmt, true);
}

/* Create and initialize a new stmt_vec_info struct for STMT.  */

stmt_vec_info
vec_info::new_stmt_vec_info (gimple *stmt)
{
  stmt_vec_info res = XCNEW (class _stmt_vec_info);
  res->stmt = stmt;

  STMT_VINFO_TYPE (res) = undef_vec_info_type;
  STMT_VINFO_RELEVANT (res) = vect_unused_in_scope;
  STMT_VINFO_VECTORIZABLE (res) = true;
  STMT_VINFO_REDUC_TYPE (res) = TREE_CODE_REDUCTION;
  STMT_VINFO_REDUC_CODE (res) = ERROR_MARK;
  STMT_VINFO_REDUC_FN (res) = IFN_LAST;
  STMT_VINFO_REDUC_IDX (res) = -1;
  STMT_VINFO_SLP_VECT_ONLY (res) = false;

  if (gimple_code (stmt) == GIMPLE_PHI
      && is_loop_header_bb_p (gimple_bb (stmt)))
    STMT_VINFO_DEF_TYPE (res) = vect_unknown_def_type;
  else
    STMT_VINFO_DEF_TYPE (res) = vect_internal_def;

  STMT_VINFO_SAME_ALIGN_REFS (res).create (0);
  STMT_SLP_TYPE (res) = loop_vect;

  /* This is really "uninitialized" until vect_compute_data_ref_alignment.  */
  res->dr_aux.misalignment = DR_MISALIGNMENT_UNINITIALIZED;

  return res;
}

/* Associate STMT with INFO.  */

void
vec_info::set_vinfo_for_stmt (gimple *stmt, stmt_vec_info info)
{
  unsigned int uid = gimple_uid (stmt);
  if (uid == 0)
    {
      gcc_checking_assert (info);
      uid = stmt_vec_infos.length () + 1;
      gimple_set_uid (stmt, uid);
      stmt_vec_infos.safe_push (info);
    }
  else
    {
      gcc_checking_assert (info == NULL);
      stmt_vec_infos[uid - 1] = info;
    }
}

/* Free the contents of stmt_vec_infos.  */

void
vec_info::free_stmt_vec_infos (void)
{
  unsigned int i;
  stmt_vec_info info;
  FOR_EACH_VEC_ELT (stmt_vec_infos, i, info)
    if (info != NULL)
      free_stmt_vec_info (info);
  stmt_vec_infos.release ();
}

/* Free STMT_INFO.  */

void
vec_info::free_stmt_vec_info (stmt_vec_info stmt_info)
{
  if (stmt_info->pattern_stmt_p)
    {
      gimple_set_bb (stmt_info->stmt, NULL);
      tree lhs = gimple_get_lhs (stmt_info->stmt);
      if (lhs && TREE_CODE (lhs) == SSA_NAME)
	release_ssa_name (lhs);
    }

  STMT_VINFO_SAME_ALIGN_REFS (stmt_info).release ();
  STMT_VINFO_SIMD_CLONE_INFO (stmt_info).release ();
  free (stmt_info);
}

/* A helper function to free scev and LOOP niter information, as well as
   clear loop constraint LOOP_C_FINITE.  */

void
vect_free_loop_info_assumptions (class loop *loop)
{
  scev_reset_htab ();
  /* We need to explicitly reset upper bound information since they are
     used even after free_numbers_of_iterations_estimates.  */
  loop->any_upper_bound = false;
  loop->any_likely_upper_bound = false;
  free_numbers_of_iterations_estimates (loop);
  loop_constraint_clear (loop, LOOP_C_FINITE);
}

/* If LOOP has been versioned during ifcvt, return the internal call
   guarding it.  */

gimple *
vect_loop_vectorized_call (class loop *loop, gcond **cond)
{
  basic_block bb = loop_preheader_edge (loop)->src;
  gimple *g;
  do
    {
      g = last_stmt (bb);
      if (g)
	break;
      if (!single_pred_p (bb))
	break;
      bb = single_pred (bb);
    }
  while (1);
  if (g && gimple_code (g) == GIMPLE_COND)
    {
      if (cond)
	*cond = as_a <gcond *> (g);
      gimple_stmt_iterator gsi = gsi_for_stmt (g);
      gsi_prev (&gsi);
      if (!gsi_end_p (gsi))
	{
	  g = gsi_stmt (gsi);
	  if (gimple_call_internal_p (g, IFN_LOOP_VECTORIZED)
	      && (tree_to_shwi (gimple_call_arg (g, 0)) == loop->num
		  || tree_to_shwi (gimple_call_arg (g, 1)) == loop->num))
	    return g;
	}
    }
  return NULL;
}

/* If LOOP has been versioned during loop distribution, return the gurading
   internal call.  */

static gimple *
vect_loop_dist_alias_call (class loop *loop)
{
  basic_block bb;
  basic_block entry;
  class loop *outer, *orig;
  gimple_stmt_iterator gsi;
  gimple *g;

  if (loop->orig_loop_num == 0)
    return NULL;

  orig = get_loop (cfun, loop->orig_loop_num);
  if (orig == NULL)
    {
      /* The original loop is somehow destroyed.  Clear the information.  */
      loop->orig_loop_num = 0;
      return NULL;
    }

  if (loop != orig)
    bb = nearest_common_dominator (CDI_DOMINATORS, loop->header, orig->header);
  else
    bb = loop_preheader_edge (loop)->src;

  outer = bb->loop_father;
  entry = ENTRY_BLOCK_PTR_FOR_FN (cfun);

  /* Look upward in dominance tree.  */
  for (; bb != entry && flow_bb_inside_loop_p (outer, bb);
       bb = get_immediate_dominator (CDI_DOMINATORS, bb))
    {
      g = last_stmt (bb);
      if (g == NULL || gimple_code (g) != GIMPLE_COND)
	continue;

      gsi = gsi_for_stmt (g);
      gsi_prev (&gsi);
      if (gsi_end_p (gsi))
	continue;

      g = gsi_stmt (gsi);
      /* The guarding internal function call must have the same distribution
	 alias id.  */
      if (gimple_call_internal_p (g, IFN_LOOP_DIST_ALIAS)
	  && (tree_to_shwi (gimple_call_arg (g, 0)) == loop->orig_loop_num))
	return g;
    }
  return NULL;
}

/* Set the uids of all the statements in basic blocks inside loop
   represented by LOOP_VINFO. LOOP_VECTORIZED_CALL is the internal
   call guarding the loop which has been if converted.  */
static void
set_uid_loop_bbs (loop_vec_info loop_vinfo, gimple *loop_vectorized_call)
{
  tree arg = gimple_call_arg (loop_vectorized_call, 1);
  basic_block *bbs;
  unsigned int i;
  class loop *scalar_loop = get_loop (cfun, tree_to_shwi (arg));

  LOOP_VINFO_SCALAR_LOOP (loop_vinfo) = scalar_loop;
  gcc_checking_assert (vect_loop_vectorized_call (scalar_loop)
		       == loop_vectorized_call);
  /* If we are going to vectorize outer loop, prevent vectorization
     of the inner loop in the scalar loop - either the scalar loop is
     thrown away, so it is a wasted work, or is used only for
     a few iterations.  */
  if (scalar_loop->inner)
    {
      gimple *g = vect_loop_vectorized_call (scalar_loop->inner);
      if (g)
	{
	  arg = gimple_call_arg (g, 0);
	  get_loop (cfun, tree_to_shwi (arg))->dont_vectorize = true;
	  fold_loop_internal_call (g, boolean_false_node);
	}
    }
  bbs = get_loop_body (scalar_loop);
  for (i = 0; i < scalar_loop->num_nodes; i++)
    {
      basic_block bb = bbs[i];
      gimple_stmt_iterator gsi;
      for (gsi = gsi_start_phis (bb); !gsi_end_p (gsi); gsi_next (&gsi))
	{
	  gimple *phi = gsi_stmt (gsi);
	  gimple_set_uid (phi, 0);
	}
      for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi))
	{
	  gimple *stmt = gsi_stmt (gsi);
	  gimple_set_uid (stmt, 0);
	}
    }
  free (bbs);
}

/* Try to vectorize LOOP.  */

static unsigned
try_vectorize_loop_1 (hash_table<simduid_to_vf> *&simduid_to_vf_htab,
		      unsigned *num_vectorized_loops, loop_p loop,
		      gimple *loop_vectorized_call,
		      gimple *loop_dist_alias_call)
{
  unsigned ret = 0;
  vec_info_shared shared;
  auto_purge_vect_location sentinel;
  vect_location = find_loop_location (loop);

  if (LOCATION_LOCUS (vect_location.get_location_t ()) != UNKNOWN_LOCATION
      && dump_enabled_p ())
    dump_printf (MSG_NOTE | MSG_PRIORITY_INTERNALS,
		 "\nAnalyzing loop at %s:%d\n",
		 LOCATION_FILE (vect_location.get_location_t ()),
		 LOCATION_LINE (vect_location.get_location_t ()));

  opt_loop_vec_info loop_vinfo = opt_loop_vec_info::success (NULL);
  /* In the case of epilogue vectorization the loop already has its
     loop_vec_info set, we do not require to analyze the loop in this case.  */
  if (loop_vec_info vinfo = loop_vec_info_for_loop (loop))
    loop_vinfo = opt_loop_vec_info::success (vinfo);
  else
    {
      /* Try to analyze the loop, retaining an opt_problem if dump_enabled_p.  */
      loop_vinfo = vect_analyze_loop (loop, &shared);
      loop->aux = loop_vinfo;
    }

  if (!loop_vinfo)
    if (dump_enabled_p ())
      if (opt_problem *problem = loop_vinfo.get_problem ())
	{
	  dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			   "couldn't vectorize loop\n");
	  problem->emit_and_clear ();
	}

  if (!loop_vinfo || !LOOP_VINFO_VECTORIZABLE_P (loop_vinfo))
    {
      /* Free existing information if loop is analyzed with some
	 assumptions.  */
      if (loop_constraint_set_p (loop, LOOP_C_FINITE))
	vect_free_loop_info_assumptions (loop);

      /* If we applied if-conversion then try to vectorize the
	 BB of innermost loops.
	 ???  Ideally BB vectorization would learn to vectorize
	 control flow by applying if-conversion on-the-fly, the
	 following retains the if-converted loop body even when
	 only non-if-converted parts took part in BB vectorization.  */
      if (flag_tree_slp_vectorize != 0
	  && loop_vectorized_call
	  && ! loop->inner)
	{
	  basic_block bb = loop->header;
	  bool require_loop_vectorize = false;
	  for (gimple_stmt_iterator gsi = gsi_start_bb (bb);
	       !gsi_end_p (gsi); gsi_next (&gsi))
	    {
	      gimple *stmt = gsi_stmt (gsi);
	      gcall *call = dyn_cast <gcall *> (stmt);
	      if (call && gimple_call_internal_p (call))
		{
		  internal_fn ifn = gimple_call_internal_fn (call);
		  if (ifn == IFN_MASK_LOAD || ifn == IFN_MASK_STORE
		      /* Don't keep the if-converted parts when the ifn with
			 specifc type is not supported by the backend.  */
		      || (direct_internal_fn_p (ifn)
			  && !direct_internal_fn_supported_p
			  (call, OPTIMIZE_FOR_SPEED)))
		    {
		      require_loop_vectorize = true;
		      break;
		    }
		}
	      gimple_set_uid (stmt, -1);
	      gimple_set_visited (stmt, false);
	    }
	  if (!require_loop_vectorize && vect_slp_bb (bb))
	    {
	      if (dump_enabled_p ())
		dump_printf_loc (MSG_NOTE, vect_location,
				 "basic block vectorized\n");
	      fold_loop_internal_call (loop_vectorized_call,
				       boolean_true_node);
	      loop_vectorized_call = NULL;
	      ret |= TODO_cleanup_cfg | TODO_update_ssa_only_virtuals;
	    }
	}
      /* If outer loop vectorization fails for LOOP_VECTORIZED guarded
	 loop, don't vectorize its inner loop; we'll attempt to
	 vectorize LOOP_VECTORIZED guarded inner loop of the scalar
	 loop version.  */
      if (loop_vectorized_call && loop->inner)
	loop->inner->dont_vectorize = true;
      return ret;
    }

  if (!dbg_cnt (vect_loop))
    {
      /* Free existing information if loop is analyzed with some
	 assumptions.  */
      if (loop_constraint_set_p (loop, LOOP_C_FINITE))
	vect_free_loop_info_assumptions (loop);
      return ret;
    }

  if (loop_vectorized_call)
    set_uid_loop_bbs (loop_vinfo, loop_vectorized_call);

  unsigned HOST_WIDE_INT bytes;
  if (dump_enabled_p ())
    {
      if (GET_MODE_SIZE (loop_vinfo->vector_mode).is_constant (&bytes))
	dump_printf_loc (MSG_OPTIMIZED_LOCATIONS, vect_location,
			 "loop vectorized using %wu byte vectors\n", bytes);
      else
	dump_printf_loc (MSG_OPTIMIZED_LOCATIONS, vect_location,
			 "loop vectorized using variable length vectors\n");
    }

  loop_p new_loop = vect_transform_loop (loop_vinfo,
					 loop_vectorized_call);
  (*num_vectorized_loops)++;
  /* Now that the loop has been vectorized, allow it to be unrolled
     etc.  */
  loop->force_vectorize = false;

  if (loop->simduid)
    {
      simduid_to_vf *simduid_to_vf_data = XNEW (simduid_to_vf);
      if (!simduid_to_vf_htab)
	simduid_to_vf_htab = new hash_table<simduid_to_vf> (15);
      simduid_to_vf_data->simduid = DECL_UID (loop->simduid);
      simduid_to_vf_data->vf = loop_vinfo->vectorization_factor;
      *simduid_to_vf_htab->find_slot (simduid_to_vf_data, INSERT)
	  = simduid_to_vf_data;
    }

  if (loop_vectorized_call)
    {
      fold_loop_internal_call (loop_vectorized_call, boolean_true_node);
      loop_vectorized_call = NULL;
      ret |= TODO_cleanup_cfg;
    }
  if (loop_dist_alias_call)
    {
      tree value = gimple_call_arg (loop_dist_alias_call, 1);
      fold_loop_internal_call (loop_dist_alias_call, value);
      loop_dist_alias_call = NULL;
      ret |= TODO_cleanup_cfg;
    }

  /* Epilogue of vectorized loop must be vectorized too.  */
  if (new_loop)
    {
      /* Don't include vectorized epilogues in the "vectorized loops" count.
       */
      unsigned dont_count = *num_vectorized_loops;
      ret |= try_vectorize_loop_1 (simduid_to_vf_htab, &dont_count,
				   new_loop, NULL, NULL);
    }

  return ret;
}

/* Try to vectorize LOOP.  */

static unsigned
try_vectorize_loop (hash_table<simduid_to_vf> *&simduid_to_vf_htab,
		    unsigned *num_vectorized_loops, loop_p loop)
{
  if (!((flag_tree_loop_vectorize
	 && optimize_loop_nest_for_speed_p (loop))
	|| loop->force_vectorize))
    return 0;

  return try_vectorize_loop_1 (simduid_to_vf_htab, num_vectorized_loops, loop,
			       vect_loop_vectorized_call (loop),
			       vect_loop_dist_alias_call (loop));
}


/* Function vectorize_loops.

   Entry point to loop vectorization phase.  */

unsigned
vectorize_loops (void)
{
  unsigned int i;
  unsigned int num_vectorized_loops = 0;
  unsigned int vect_loops_num;
  class loop *loop;
  hash_table<simduid_to_vf> *simduid_to_vf_htab = NULL;
  hash_table<simd_array_to_simduid> *simd_array_to_simduid_htab = NULL;
  bool any_ifcvt_loops = false;
  unsigned ret = 0;

  vect_loops_num = number_of_loops (cfun);

  /* Bail out if there are no loops.  */
  if (vect_loops_num <= 1)
    return 0;

  if (cfun->has_simduid_loops)
    note_simd_array_uses (&simd_array_to_simduid_htab);

  /*  ----------- Analyze loops. -----------  */

  /* If some loop was duplicated, it gets bigger number
     than all previously defined loops.  This fact allows us to run
     only over initial loops skipping newly generated ones.  */
  FOR_EACH_LOOP (loop, 0)
    if (loop->dont_vectorize)
      {
	any_ifcvt_loops = true;
	/* If-conversion sometimes versions both the outer loop
	   (for the case when outer loop vectorization might be
	   desirable) as well as the inner loop in the scalar version
	   of the loop.  So we have:
	    if (LOOP_VECTORIZED (1, 3))
	      {
		loop1
		  loop2
	      }
	    else
	      loop3 (copy of loop1)
		if (LOOP_VECTORIZED (4, 5))
		  loop4 (copy of loop2)
		else
		  loop5 (copy of loop4)
	   If FOR_EACH_LOOP gives us loop3 first (which has
	   dont_vectorize set), make sure to process loop1 before loop4;
	   so that we can prevent vectorization of loop4 if loop1
	   is successfully vectorized.  */
	if (loop->inner)
	  {
	    gimple *loop_vectorized_call
	      = vect_loop_vectorized_call (loop);
	    if (loop_vectorized_call
		&& vect_loop_vectorized_call (loop->inner))
	      {
		tree arg = gimple_call_arg (loop_vectorized_call, 0);
		class loop *vector_loop
		  = get_loop (cfun, tree_to_shwi (arg));
		if (vector_loop && vector_loop != loop)
		  {
		    /* Make sure we don't vectorize it twice.  */
		    vector_loop->dont_vectorize = true;
		    ret |= try_vectorize_loop (simduid_to_vf_htab,
					       &num_vectorized_loops,
					       vector_loop);
		  }
	      }
	  }
      }
    else
      ret |= try_vectorize_loop (simduid_to_vf_htab, &num_vectorized_loops,
				 loop);

  vect_location = dump_user_location_t ();

  statistics_counter_event (cfun, "Vectorized loops", num_vectorized_loops);
  if (dump_enabled_p ()
      || (num_vectorized_loops > 0 && dump_enabled_p ()))
    dump_printf_loc (MSG_NOTE, vect_location,
                     "vectorized %u loops in function.\n",
                     num_vectorized_loops);

  /*  ----------- Finalize. -----------  */

  if (any_ifcvt_loops)
    for (i = 1; i < number_of_loops (cfun); i++)
      {
	loop = get_loop (cfun, i);
	if (loop && loop->dont_vectorize)
	  {
	    gimple *g = vect_loop_vectorized_call (loop);
	    if (g)
	      {
		fold_loop_internal_call (g, boolean_false_node);
		ret |= TODO_cleanup_cfg;
		g = NULL;
	      }
	    else
	      g = vect_loop_dist_alias_call (loop);

	    if (g)
	      {
		fold_loop_internal_call (g, boolean_false_node);
		ret |= TODO_cleanup_cfg;
	      }
	  }
      }

  for (i = 1; i < number_of_loops (cfun); i++)
    {
      loop_vec_info loop_vinfo;
      bool has_mask_store;

      loop = get_loop (cfun, i);
      if (!loop || !loop->aux)
	continue;
      loop_vinfo = (loop_vec_info) loop->aux;
      has_mask_store = LOOP_VINFO_HAS_MASK_STORE (loop_vinfo);
      delete loop_vinfo;
      if (has_mask_store
	  && targetm.vectorize.empty_mask_is_expensive (IFN_MASK_STORE))
	optimize_mask_stores (loop);
      loop->aux = NULL;
    }

  /* Fold IFN_GOMP_SIMD_{VF,LANE,LAST_LANE,ORDERED_{START,END}} builtins.  */
  if (cfun->has_simduid_loops)
    adjust_simduid_builtins (simduid_to_vf_htab);

  /* Shrink any "omp array simd" temporary arrays to the
     actual vectorization factors.  */
  if (simd_array_to_simduid_htab)
    shrink_simd_arrays (simd_array_to_simduid_htab, simduid_to_vf_htab);
  delete simduid_to_vf_htab;
  cfun->has_simduid_loops = false;

  if (num_vectorized_loops > 0)
    {
      /* If we vectorized any loop only virtual SSA form needs to be updated.
	 ???  Also while we try hard to update loop-closed SSA form we fail
	 to properly do this in some corner-cases (see PR56286).  */
      rewrite_into_loop_closed_ssa (NULL, TODO_update_ssa_only_virtuals);
      return TODO_cleanup_cfg;
    }

  return ret;
}


/* Entry point to the simduid cleanup pass.  */

namespace {

const pass_data pass_data_simduid_cleanup =
{
  GIMPLE_PASS, /* type */
  "simduid", /* name */
  OPTGROUP_NONE, /* optinfo_flags */
  TV_NONE, /* tv_id */
  ( PROP_ssa | PROP_cfg ), /* properties_required */
  0, /* properties_provided */
  0, /* properties_destroyed */
  0, /* todo_flags_start */
  0, /* todo_flags_finish */
};

class pass_simduid_cleanup : public gimple_opt_pass
{
public:
  pass_simduid_cleanup (gcc::context *ctxt)
    : gimple_opt_pass (pass_data_simduid_cleanup, ctxt)
  {}

  /* opt_pass methods: */
  opt_pass * clone () { return new pass_simduid_cleanup (m_ctxt); }
  virtual bool gate (function *fun) { return fun->has_simduid_loops; }
  virtual unsigned int execute (function *);

}; // class pass_simduid_cleanup

unsigned int
pass_simduid_cleanup::execute (function *fun)
{
  hash_table<simd_array_to_simduid> *simd_array_to_simduid_htab = NULL;

  note_simd_array_uses (&simd_array_to_simduid_htab);

  /* Fold IFN_GOMP_SIMD_{VF,LANE,LAST_LANE,ORDERED_{START,END}} builtins.  */
  adjust_simduid_builtins (NULL);

  /* Shrink any "omp array simd" temporary arrays to the
     actual vectorization factors.  */
  if (simd_array_to_simduid_htab)
    shrink_simd_arrays (simd_array_to_simduid_htab, NULL);
  fun->has_simduid_loops = false;
  return 0;
}

}  // anon namespace

gimple_opt_pass *
make_pass_simduid_cleanup (gcc::context *ctxt)
{
  return new pass_simduid_cleanup (ctxt);
}


/*  Entry point to basic block SLP phase.  */

namespace {

const pass_data pass_data_slp_vectorize =
{
  GIMPLE_PASS, /* type */
  "slp", /* name */
  OPTGROUP_LOOP | OPTGROUP_VEC, /* optinfo_flags */
  TV_TREE_SLP_VECTORIZATION, /* tv_id */
  ( PROP_ssa | PROP_cfg ), /* properties_required */
  0, /* properties_provided */
  0, /* properties_destroyed */
  0, /* todo_flags_start */
  TODO_update_ssa, /* todo_flags_finish */
};

class pass_slp_vectorize : public gimple_opt_pass
{
public:
  pass_slp_vectorize (gcc::context *ctxt)
    : gimple_opt_pass (pass_data_slp_vectorize, ctxt)
  {}

  /* opt_pass methods: */
  opt_pass * clone () { return new pass_slp_vectorize (m_ctxt); }
  virtual bool gate (function *) { return flag_tree_slp_vectorize != 0; }
  virtual unsigned int execute (function *);

}; // class pass_slp_vectorize

unsigned int
pass_slp_vectorize::execute (function *fun)
{
  auto_purge_vect_location sentinel;
  basic_block bb;

  bool in_loop_pipeline = scev_initialized_p ();
  if (!in_loop_pipeline)
    {
      loop_optimizer_init (LOOPS_NORMAL);
      scev_initialize ();
    }

  /* Mark all stmts as not belonging to the current region and unvisited.  */
  FOR_EACH_BB_FN (bb, fun)
    {
      for (gimple_stmt_iterator gsi = gsi_start_bb (bb); !gsi_end_p (gsi);
	   gsi_next (&gsi))
	{
	  gimple *stmt = gsi_stmt (gsi);
	  gimple_set_uid (stmt, -1);
	  gimple_set_visited (stmt, false);
	}
    }

  FOR_EACH_BB_FN (bb, fun)
    {
      if (vect_slp_bb (bb))
	if (dump_enabled_p ())
	  dump_printf_loc (MSG_NOTE, vect_location, "basic block vectorized\n");
    }

  if (!in_loop_pipeline)
    {
      scev_finalize ();
      loop_optimizer_finalize ();
    }

  return 0;
}

} // anon namespace

gimple_opt_pass *
make_pass_slp_vectorize (gcc::context *ctxt)
{
  return new pass_slp_vectorize (ctxt);
}


/* Increase alignment of global arrays to improve vectorization potential.
   TODO:
   - Consider also structs that have an array field.
   - Use ipa analysis to prune arrays that can't be vectorized?
     This should involve global alignment analysis and in the future also
     array padding.  */

static unsigned get_vec_alignment_for_type (tree);
static hash_map<tree, unsigned> *type_align_map;

/* Return alignment of array's vector type corresponding to scalar type.
   0 if no vector type exists.  */
static unsigned
get_vec_alignment_for_array_type (tree type) 
{
  gcc_assert (TREE_CODE (type) == ARRAY_TYPE);
  poly_uint64 array_size, vector_size;

  tree scalar_type = strip_array_types (type);
  tree vectype = get_related_vectype_for_scalar_type (VOIDmode, scalar_type);
  if (!vectype
      || !poly_int_tree_p (TYPE_SIZE (type), &array_size)
      || !poly_int_tree_p (TYPE_SIZE (vectype), &vector_size)
      || maybe_lt (array_size, vector_size))
    return 0;

  return TYPE_ALIGN (vectype);
}

/* Return alignment of field having maximum alignment of vector type
   corresponding to it's scalar type. For now, we only consider fields whose
   offset is a multiple of it's vector alignment.
   0 if no suitable field is found.  */
static unsigned
get_vec_alignment_for_record_type (tree type) 
{
  gcc_assert (TREE_CODE (type) == RECORD_TYPE);

  unsigned max_align = 0, alignment;
  HOST_WIDE_INT offset;
  tree offset_tree;

  if (TYPE_PACKED (type))
    return 0;

  unsigned *slot = type_align_map->get (type);
  if (slot)
    return *slot;

  for (tree field = first_field (type);
       field != NULL_TREE;
       field = DECL_CHAIN (field))
    {
      /* Skip if not FIELD_DECL or if alignment is set by user.  */ 
      if (TREE_CODE (field) != FIELD_DECL
	  || DECL_USER_ALIGN (field)
	  || DECL_ARTIFICIAL (field))
	continue;

      /* We don't need to process the type further if offset is variable,
	 since the offsets of remaining members will also be variable.  */
      if (TREE_CODE (DECL_FIELD_OFFSET (field)) != INTEGER_CST
	  || TREE_CODE (DECL_FIELD_BIT_OFFSET (field)) != INTEGER_CST)
	break;

      /* Similarly stop processing the type if offset_tree
	 does not fit in unsigned HOST_WIDE_INT.  */
      offset_tree = bit_position (field);
      if (!tree_fits_uhwi_p (offset_tree))
	break;

      offset = tree_to_uhwi (offset_tree); 
      alignment = get_vec_alignment_for_type (TREE_TYPE (field));

      /* Get maximum alignment of vectorized field/array among those members
	 whose offset is multiple of the vector alignment.  */ 
      if (alignment
	  && (offset % alignment == 0)
	  && (alignment > max_align))
	max_align = alignment;
    }

  type_align_map->put (type, max_align);
  return max_align;
}

/* Return alignment of vector type corresponding to decl's scalar type
   or 0 if it doesn't exist or the vector alignment is lesser than
   decl's alignment.  */
static unsigned
get_vec_alignment_for_type (tree type)
{
  if (type == NULL_TREE)
    return 0;

  gcc_assert (TYPE_P (type));

  static unsigned alignment = 0;
  switch (TREE_CODE (type))
    {
      case ARRAY_TYPE:
	alignment = get_vec_alignment_for_array_type (type);
	break;
      case RECORD_TYPE:
	alignment = get_vec_alignment_for_record_type (type);
	break;
      default:
	alignment = 0;
	break;
    }

  return (alignment > TYPE_ALIGN (type)) ? alignment : 0;
}

/* Entry point to increase_alignment pass.  */
static unsigned int
increase_alignment (void)
{
  varpool_node *vnode;

  vect_location = dump_user_location_t ();
  type_align_map = new hash_map<tree, unsigned>;

  /* Increase the alignment of all global arrays for vectorization.  */
  FOR_EACH_DEFINED_VARIABLE (vnode)
    {
      tree decl = vnode->decl;
      unsigned int alignment;

      if ((decl_in_symtab_p (decl)
	  && !symtab_node::get (decl)->can_increase_alignment_p ())
	  || DECL_USER_ALIGN (decl) || DECL_ARTIFICIAL (decl))
	continue;

      alignment = get_vec_alignment_for_type (TREE_TYPE (decl));
      if (alignment && vect_can_force_dr_alignment_p (decl, alignment))
        {
	  vnode->increase_alignment (alignment);
	  if (dump_enabled_p ())
	    dump_printf (MSG_NOTE, "Increasing alignment of decl: %T\n", decl);
        }
    }

  delete type_align_map;
  return 0;
}


namespace {

const pass_data pass_data_ipa_increase_alignment =
{
  SIMPLE_IPA_PASS, /* type */
  "increase_alignment", /* name */
  OPTGROUP_LOOP | OPTGROUP_VEC, /* optinfo_flags */
  TV_IPA_OPT, /* tv_id */
  0, /* properties_required */
  0, /* properties_provided */
  0, /* properties_destroyed */
  0, /* todo_flags_start */
  0, /* todo_flags_finish */
};

class pass_ipa_increase_alignment : public simple_ipa_opt_pass
{
public:
  pass_ipa_increase_alignment (gcc::context *ctxt)
    : simple_ipa_opt_pass (pass_data_ipa_increase_alignment, ctxt)
  {}

  /* opt_pass methods: */
  virtual bool gate (function *)
    {
      return flag_section_anchors && flag_tree_loop_vectorize;
    }

  virtual unsigned int execute (function *) { return increase_alignment (); }

}; // class pass_ipa_increase_alignment

} // anon namespace

simple_ipa_opt_pass *
make_pass_ipa_increase_alignment (gcc::context *ctxt)
{
  return new pass_ipa_increase_alignment (ctxt);
}

/* If the condition represented by T is a comparison or the SSA name
   result of a comparison, extract the comparison's operands.  Represent
   T as NE_EXPR <T, 0> otherwise.  */

void
scalar_cond_masked_key::get_cond_ops_from_tree (tree t)
{
  if (TREE_CODE_CLASS (TREE_CODE (t)) == tcc_comparison)
    {
      this->code = TREE_CODE (t);
      this->op0 = TREE_OPERAND (t, 0);
      this->op1 = TREE_OPERAND (t, 1);
      return;
    }

  if (TREE_CODE (t) == SSA_NAME)
    if (gassign *stmt = dyn_cast<gassign *> (SSA_NAME_DEF_STMT (t)))
      {
	tree_code code = gimple_assign_rhs_code (stmt);
	if (TREE_CODE_CLASS (code) == tcc_comparison)
	  {
	    this->code = code;
	    this->op0 = gimple_assign_rhs1 (stmt);
	    this->op1 = gimple_assign_rhs2 (stmt);
	    return;
	  }
      }

  this->code = NE_EXPR;
  this->op0 = t;
  this->op1 = build_zero_cst (TREE_TYPE (t));
}