aboutsummaryrefslogtreecommitdiff
path: root/gcc/cppinit.c
blob: 351ce187eac5b4bf5ff874984df40f14fa9f7f68 (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
/* CPP Library.
   Copyright (C) 1986, 1987, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
   1999, 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
   Contributed by Per Bothner, 1994-95.
   Based on CCCP program by Paul Rubin, June 1986
   Adapted to ANSI C, Richard Stallman, Jan 1987

This program 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 2, or (at your option) any
later version.

This program 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 this program; if not, write to the Free Software
Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */

#include "config.h"
#include "system.h"
#include "cpplib.h"
#include "cpphash.h"
#include "mkdeps.h"

static void init_library (void);
static void mark_named_operators (cpp_reader *);
static void read_original_filename (cpp_reader *);
static void read_original_directory (cpp_reader *);
static void post_options (cpp_reader *);

/* If we have designated initializers (GCC >2.7) these tables can be
   initialized, constant data.  Otherwise, they have to be filled in at
   runtime.  */
#if HAVE_DESIGNATED_INITIALIZERS

#define init_trigraph_map()  /* Nothing.  */
#define TRIGRAPH_MAP \
__extension__ const uchar _cpp_trigraph_map[UCHAR_MAX + 1] = {

#define END };
#define s(p, v) [p] = v,

#else

#define TRIGRAPH_MAP uchar _cpp_trigraph_map[UCHAR_MAX + 1] = { 0 }; \
 static void init_trigraph_map (void) { \
 unsigned char *x = _cpp_trigraph_map;

#define END }
#define s(p, v) x[p] = v;

#endif

TRIGRAPH_MAP
  s('=', '#')	s(')', ']')	s('!', '|')
  s('(', '[')	s('\'', '^')	s('>', '}')
  s('/', '\\')	s('<', '{')	s('-', '~')
END

#undef s
#undef END
#undef TRIGRAPH_MAP

/* A set of booleans indicating what CPP features each source language
   requires.  */
struct lang_flags
{
  char c99;
  char cplusplus;
  char extended_numbers;
  char std;
  char cplusplus_comments;
  char digraphs;
};

static const struct lang_flags lang_defaults[] =
{ /*              c99 c++ xnum std  //   digr  */
  /* GNUC89 */  { 0,  0,  1,   0,   1,   1     },
  /* GNUC99 */  { 1,  0,  1,   0,   1,   1     },
  /* STDC89 */  { 0,  0,  0,   1,   0,   0     },
  /* STDC94 */  { 0,  0,  0,   1,   0,   1     },
  /* STDC99 */  { 1,  0,  1,   1,   1,   1     },
  /* GNUCXX */  { 0,  1,  1,   0,   1,   1     },
  /* CXX98  */  { 0,  1,  1,   1,   1,   1     },
  /* ASM    */  { 0,  0,  1,   0,   1,   0     }
};

/* Sets internal flags correctly for a given language.  */
void
cpp_set_lang (cpp_reader *pfile, enum c_lang lang)
{
  const struct lang_flags *l = &lang_defaults[(int) lang];

  CPP_OPTION (pfile, lang) = lang;

  CPP_OPTION (pfile, c99)		 = l->c99;
  CPP_OPTION (pfile, cplusplus)		 = l->cplusplus;
  CPP_OPTION (pfile, extended_numbers)	 = l->extended_numbers;
  CPP_OPTION (pfile, std)		 = l->std;
  CPP_OPTION (pfile, trigraphs)		 = l->std;
  CPP_OPTION (pfile, cplusplus_comments) = l->cplusplus_comments;
  CPP_OPTION (pfile, digraphs)		 = l->digraphs;
}

/* Initialize library global state.  */
static void
init_library (void)
{
  static int initialized = 0;

  if (! initialized)
    {
      initialized = 1;

      /* Set up the trigraph map.  This doesn't need to do anything if
	 we were compiled with a compiler that supports C99 designated
	 initializers.  */
      init_trigraph_map ();
    }
}

/* Initialize a cpp_reader structure.  */
cpp_reader *
cpp_create_reader (enum c_lang lang, hash_table *table)
{
  cpp_reader *pfile;

  /* Initialize this instance of the library if it hasn't been already.  */
  init_library ();

  pfile = xcalloc (1, sizeof (cpp_reader));

  cpp_set_lang (pfile, lang);
  CPP_OPTION (pfile, warn_multichar) = 1;
  CPP_OPTION (pfile, discard_comments) = 1;
  CPP_OPTION (pfile, discard_comments_in_macro_exp) = 1;
  CPP_OPTION (pfile, show_column) = 1;
  CPP_OPTION (pfile, tabstop) = 8;
  CPP_OPTION (pfile, operator_names) = 1;
  CPP_OPTION (pfile, warn_trigraphs) = 2;
  CPP_OPTION (pfile, warn_endif_labels) = 1;
  CPP_OPTION (pfile, warn_deprecated) = 1;
  CPP_OPTION (pfile, warn_long_long) = !CPP_OPTION (pfile, c99);
  CPP_OPTION (pfile, dollars_in_ident) = 1;
  CPP_OPTION (pfile, warn_dollars) = 1;

  /* Default CPP arithmetic to something sensible for the host for the
     benefit of dumb users like fix-header.  */
  CPP_OPTION (pfile, precision) = CHAR_BIT * sizeof (long);
  CPP_OPTION (pfile, char_precision) = CHAR_BIT;
  CPP_OPTION (pfile, wchar_precision) = CHAR_BIT * sizeof (int);
  CPP_OPTION (pfile, int_precision) = CHAR_BIT * sizeof (int);
  CPP_OPTION (pfile, unsigned_char) = 0;
  CPP_OPTION (pfile, unsigned_wchar) = 1;
  CPP_OPTION (pfile, bytes_big_endian) = 1;  /* does not matter */

  /* Default to no charset conversion.  */
  CPP_OPTION (pfile, narrow_charset) = 0;
  CPP_OPTION (pfile, wide_charset) = 0;

  /* A fake empty "directory" used as the starting point for files
     looked up without a search path.  Name cannot be '/' because we
     don't want to prepend anything at all to filenames using it.  All
     other entries are correct zero-initialized.  */
  pfile->no_search_path.name = (char *) "";

  /* Initialize the line map.  Start at logical line 1, so we can use
     a line number of zero for special states.  */
  linemap_init (&pfile->line_maps);
  pfile->line = 1;

  /* Initialize lexer state.  */
  pfile->state.save_comments = ! CPP_OPTION (pfile, discard_comments);

  /* Set up static tokens.  */
  pfile->avoid_paste.type = CPP_PADDING;
  pfile->avoid_paste.val.source = NULL;
  pfile->eof.type = CPP_EOF;
  pfile->eof.flags = 0;

  /* Create a token buffer for the lexer.  */
  _cpp_init_tokenrun (&pfile->base_run, 250);
  pfile->cur_run = &pfile->base_run;
  pfile->cur_token = pfile->base_run.base;

  /* Initialize the base context.  */
  pfile->context = &pfile->base_context;
  pfile->base_context.macro = 0;
  pfile->base_context.prev = pfile->base_context.next = 0;

  /* Aligned and unaligned storage.  */
  pfile->a_buff = _cpp_get_buff (pfile, 0);
  pfile->u_buff = _cpp_get_buff (pfile, 0);

  /* The expression parser stack.  */
  _cpp_expand_op_stack (pfile);

  /* Initialize the buffer obstack.  */
  _obstack_begin (&pfile->buffer_ob, 0, 0,
		  (void *(*) (long)) xmalloc,
		  (void (*) (void *)) free);

  _cpp_init_files (pfile);

  _cpp_init_hashtable (pfile, table);

  return pfile;
}

/* Free resources used by PFILE.  Accessing PFILE after this function
   returns leads to undefined behavior.  Returns the error count.  */
void
cpp_destroy (cpp_reader *pfile)
{
  cpp_context *context, *contextn;
  tokenrun *run, *runn;

  free (pfile->op_stack);

  while (CPP_BUFFER (pfile) != NULL)
    _cpp_pop_buffer (pfile);

  if (pfile->out.base)
    free (pfile->out.base);

  if (pfile->macro_buffer)
    {
      free (pfile->macro_buffer);
      pfile->macro_buffer = NULL;
      pfile->macro_buffer_len = 0;
    }

  if (pfile->deps)
    deps_free (pfile->deps);
  obstack_free (&pfile->buffer_ob, 0);

  _cpp_destroy_hashtable (pfile);
  _cpp_cleanup_files (pfile);
  _cpp_destroy_iconv (pfile);

  _cpp_free_buff (pfile->a_buff);
  _cpp_free_buff (pfile->u_buff);
  _cpp_free_buff (pfile->free_buffs);

  for (run = &pfile->base_run; run; run = runn)
    {
      runn = run->next;
      free (run->base);
      if (run != &pfile->base_run)
	free (run);
    }

  for (context = pfile->base_context.next; context; context = contextn)
    {
      contextn = context->next;
      free (context);
    }

  linemap_free (&pfile->line_maps);
  free (pfile);
}

/* This structure defines one built-in identifier.  A node will be
   entered in the hash table under the name NAME, with value VALUE.

   There are two tables of these.  builtin_array holds all the
   "builtin" macros: these are handled by builtin_macro() in
   cppmacro.c.  Builtin is somewhat of a misnomer -- the property of
   interest is that these macros require special code to compute their
   expansions.  The value is a "builtin_type" enumerator.

   operator_array holds the C++ named operators.  These are keywords
   which act as aliases for punctuators.  In C++, they cannot be
   altered through #define, and #if recognizes them as operators.  In
   C, these are not entered into the hash table at all (but see
   <iso646.h>).  The value is a token-type enumerator.  */
struct builtin
{
  const uchar *name;
  unsigned short len;
  unsigned short value;
};

#define B(n, t)    { DSC(n), t }
static const struct builtin builtin_array[] =
{
  B("__TIME__",		 BT_TIME),
  B("__DATE__",		 BT_DATE),
  B("__FILE__",		 BT_FILE),
  B("__BASE_FILE__",	 BT_BASE_FILE),
  B("__LINE__",		 BT_SPECLINE),
  B("__INCLUDE_LEVEL__", BT_INCLUDE_LEVEL),
  /* Keep builtins not used for -traditional-cpp at the end, and
     update init_builtins() if any more are added.  */
  B("_Pragma",		 BT_PRAGMA),
  B("__STDC__",		 BT_STDC),
};

static const struct builtin operator_array[] =
{
  B("and",	CPP_AND_AND),
  B("and_eq",	CPP_AND_EQ),
  B("bitand",	CPP_AND),
  B("bitor",	CPP_OR),
  B("compl",	CPP_COMPL),
  B("not",	CPP_NOT),
  B("not_eq",	CPP_NOT_EQ),
  B("or",	CPP_OR_OR),
  B("or_eq",	CPP_OR_EQ),
  B("xor",	CPP_XOR),
  B("xor_eq",	CPP_XOR_EQ)
};
#undef B

/* Mark the C++ named operators in the hash table.  */
static void
mark_named_operators (cpp_reader *pfile)
{
  const struct builtin *b;

  for (b = operator_array;
       b < (operator_array + ARRAY_SIZE (operator_array));
       b++)
    {
      cpp_hashnode *hp = cpp_lookup (pfile, b->name, b->len);
      hp->flags |= NODE_OPERATOR;
      hp->is_directive = 0;
      hp->directive_index = b->value;
    }
}

/* Read the builtins table above and enter them, and language-specific
   macros, into the hash table.  HOSTED is true if this is a hosted
   environment.  */
void
cpp_init_builtins (cpp_reader *pfile, int hosted)
{
  const struct builtin *b;
  size_t n = ARRAY_SIZE (builtin_array);

  if (CPP_OPTION (pfile, traditional))
    n -= 2;

  for(b = builtin_array; b < builtin_array + n; b++)
    {
      cpp_hashnode *hp = cpp_lookup (pfile, b->name, b->len);
      hp->type = NT_MACRO;
      hp->flags |= NODE_BUILTIN | NODE_WARN;
      hp->value.builtin = b->value;
    }

  if (CPP_OPTION (pfile, cplusplus))
    _cpp_define_builtin (pfile, "__cplusplus 1");
  else if (CPP_OPTION (pfile, lang) == CLK_ASM)
    _cpp_define_builtin (pfile, "__ASSEMBLER__ 1");
  else if (CPP_OPTION (pfile, lang) == CLK_STDC94)
    _cpp_define_builtin (pfile, "__STDC_VERSION__ 199409L");
  else if (CPP_OPTION (pfile, c99))
    _cpp_define_builtin (pfile, "__STDC_VERSION__ 199901L");

  if (hosted)
    _cpp_define_builtin (pfile, "__STDC_HOSTED__ 1");
  else
    _cpp_define_builtin (pfile, "__STDC_HOSTED__ 0");

  if (CPP_OPTION (pfile, objc))
    _cpp_define_builtin (pfile, "__OBJC__ 1");
}

/* Sanity-checks are dependent on command-line options, so it is
   called as a subroutine of cpp_read_main_file ().  */
#if ENABLE_CHECKING
static void sanity_checks (cpp_reader *);
static void sanity_checks (cpp_reader *pfile)
{
  cppchar_t test = 0;
  size_t max_precision = 2 * CHAR_BIT * sizeof (cpp_num_part);

  /* Sanity checks for assumptions about CPP arithmetic and target
     type precisions made by cpplib.  */
  test--;
  if (test < 1)
    cpp_error (pfile, DL_ICE, "cppchar_t must be an unsigned type");

  if (CPP_OPTION (pfile, precision) > max_precision)
    cpp_error (pfile, DL_ICE,
	       "preprocessor arithmetic has maximum precision of %lu bits;"
	       " target requires %lu bits",
	       (unsigned long) max_precision,
	       (unsigned long) CPP_OPTION (pfile, precision));

  if (CPP_OPTION (pfile, precision) < CPP_OPTION (pfile, int_precision))
    cpp_error (pfile, DL_ICE,
	       "CPP arithmetic must be at least as precise as a target int");

  if (CPP_OPTION (pfile, char_precision) < 8)
    cpp_error (pfile, DL_ICE, "target char is less than 8 bits wide");

  if (CPP_OPTION (pfile, wchar_precision) < CPP_OPTION (pfile, char_precision))
    cpp_error (pfile, DL_ICE,
	       "target wchar_t is narrower than target char");

  if (CPP_OPTION (pfile, int_precision) < CPP_OPTION (pfile, char_precision))
    cpp_error (pfile, DL_ICE,
	       "target int is narrower than target char");

  /* This is assumed in eval_token() and could be fixed if necessary.  */
  if (sizeof (cppchar_t) > sizeof (cpp_num_part))
    cpp_error (pfile, DL_ICE, "CPP half-integer narrower than CPP character");

  if (CPP_OPTION (pfile, wchar_precision) > BITS_PER_CPPCHAR_T)
    cpp_error (pfile, DL_ICE,
	       "CPP on this host cannot handle wide character constants over"
	       " %lu bits, but the target requires %lu bits",
	       (unsigned long) BITS_PER_CPPCHAR_T,
	       (unsigned long) CPP_OPTION (pfile, wchar_precision));
}
#else
# define sanity_checks(PFILE)
#endif

/* Add a dependency target.  Can be called any number of times before
   cpp_read_main_file().  If no targets have been added before
   cpp_read_main_file(), then the default target is used.  */
void
cpp_add_dependency_target (cpp_reader *pfile, const char *target, int quote)
{
  if (!pfile->deps)
    pfile->deps = deps_init ();

  deps_add_target (pfile->deps, target, quote);
}

/* This is called after options have been parsed, and partially
   processed.  */
void
cpp_post_options (cpp_reader *pfile)
{
  sanity_checks (pfile);

  post_options (pfile);

  /* Mark named operators before handling command line macros.  */
  if (CPP_OPTION (pfile, cplusplus) && CPP_OPTION (pfile, operator_names))
    mark_named_operators (pfile);
}

/* Setup for processing input from the file named FNAME,
   or stdin if it is the empty string.  Return the original filename
   on success (e.g. foo.i->foo.c), or NULL on failure.  */
const char *
cpp_read_main_file (cpp_reader *pfile, const char *fname)
{
  if (CPP_OPTION (pfile, deps.style) != DEPS_NONE)
    {
      if (!pfile->deps)
	pfile->deps = deps_init ();

      /* Set the default target (if there is none already).  */
      deps_add_default_target (pfile->deps, fname);
    }

  if (!_cpp_stack_file (pfile, fname))
    return NULL;

  /* Set this here so the client can change the option if it wishes,
     and after stacking the main file so we don't trace the main
     file.  */
  pfile->line_maps.trace_includes = CPP_OPTION (pfile, print_include_names);

  /* For foo.i, read the original filename foo.c now, for the benefit
     of the front ends.  */
  if (CPP_OPTION (pfile, preprocessed))
    read_original_filename (pfile);

  if (CPP_OPTION (pfile, working_directory))
    {
      const char *name = pfile->map->to_file;
      const char *dir = getpwd ();
      char *dir_with_slashes = alloca (strlen (dir) + 3);

      memcpy (dir_with_slashes, dir, strlen (dir));
      memcpy (dir_with_slashes + strlen (dir), "//", 3);

      if (pfile->cb.dir_change)
	pfile->cb.dir_change (pfile, dir);
      /* Emit file renames that will be recognized by
	 read_directory_filename, since dir_change doesn't output
	 anything.  */
      _cpp_do_file_change (pfile, LC_RENAME, dir_with_slashes, 1, 0);
      _cpp_do_file_change (pfile, LC_RENAME, name, 1, 0);
    }

  return pfile->map->to_file;
}

/* For preprocessed files, if the first tokens are of the form # NUM.
   handle the directive so we know the original file name.  This will
   generate file_change callbacks, which the front ends must handle
   appropriately given their state of initialization.  */
static void
read_original_filename (cpp_reader *pfile)
{
  const cpp_token *token, *token1;

  /* Lex ahead; if the first tokens are of the form # NUM, then
     process the directive, otherwise back up.  */
  token = _cpp_lex_direct (pfile);
  if (token->type == CPP_HASH)
    {
      token1 = _cpp_lex_direct (pfile);
      _cpp_backup_tokens (pfile, 1);

      /* If it's a #line directive, handle it.  */
      if (token1->type == CPP_NUMBER)
	{
	  _cpp_handle_directive (pfile, token->flags & PREV_WHITE);
	  read_original_directory (pfile);
	  return;
	}
    }

  /* Backup as if nothing happened.  */
  _cpp_backup_tokens (pfile, 1);
}

/* For preprocessed files, if the tokens following the first filename
   line is of the form # <line> "/path/name//", handle the
   directive so we know the original current directory.  */
static void
read_original_directory (cpp_reader *pfile)
{
  const cpp_token *hash, *token;

  /* Lex ahead; if the first tokens are of the form # NUM, then
     process the directive, otherwise back up.  */
  hash = _cpp_lex_direct (pfile);
  if (hash->type != CPP_HASH)
    {
      _cpp_backup_tokens (pfile, 1);
      return;
    }

  token = _cpp_lex_direct (pfile);

  if (token->type != CPP_NUMBER)
    {
      _cpp_backup_tokens (pfile, 2);
      return;
    }

  token = _cpp_lex_direct (pfile);

  if (token->type != CPP_STRING
      || ! (token->val.str.len >= 5
	    && token->val.str.text[token->val.str.len-2] == '/'
	    && token->val.str.text[token->val.str.len-3] == '/'))
    {
      _cpp_backup_tokens (pfile, 3);
      return;
    }

  if (pfile->cb.dir_change)
    {
      char *debugdir = alloca (token->val.str.len - 3);

      memcpy (debugdir, (const char *) token->val.str.text + 1,
	      token->val.str.len - 4);
      debugdir[token->val.str.len - 4] = '\0';

      pfile->cb.dir_change (pfile, debugdir);
    }      

  /* We want to process the fake line changes as regular changes, to
     get them output.  */
  _cpp_backup_tokens (pfile, 3);

  CPP_OPTION (pfile, working_directory) = false;
}

/* This is called at the end of preprocessing.  It pops the last
   buffer and writes dependency output, and returns the number of
   errors.
 
   Maybe it should also reset state, such that you could call
   cpp_start_read with a new filename to restart processing.  */
int
cpp_finish (cpp_reader *pfile, FILE *deps_stream)
{
  /* Warn about unused macros before popping the final buffer.  */
  if (CPP_OPTION (pfile, warn_unused_macros))
    cpp_forall_identifiers (pfile, _cpp_warn_if_unused_macro, NULL);

  /* cpplex.c leaves the final buffer on the stack.  This it so that
     it returns an unending stream of CPP_EOFs to the client.  If we
     popped the buffer, we'd dereference a NULL buffer pointer and
     segfault.  It's nice to allow the client to do worry-free excess
     cpp_get_token calls.  */
  while (pfile->buffer)
    _cpp_pop_buffer (pfile);

  /* Don't write the deps file if there are errors.  */
  if (CPP_OPTION (pfile, deps.style) != DEPS_NONE
      && deps_stream && pfile->errors == 0)
    {
      deps_write (pfile->deps, deps_stream, 72);

      if (CPP_OPTION (pfile, deps.phony_targets))
	deps_phony_targets (pfile->deps, deps_stream);
    }

  /* Report on headers that could use multiple include guards.  */
  if (CPP_OPTION (pfile, print_include_names))
    _cpp_report_missing_guards (pfile);

  return pfile->errors;
}

static void
post_options (cpp_reader *pfile)
{
  /* -Wtraditional is not useful in C++ mode.  */
  if (CPP_OPTION (pfile, cplusplus))
    CPP_OPTION (pfile, warn_traditional) = 0;

  /* Permanently disable macro expansion if we are rescanning
     preprocessed text.  Read preprocesed source in ISO mode.  */
  if (CPP_OPTION (pfile, preprocessed))
    {
      pfile->state.prevent_expansion = 1;
      CPP_OPTION (pfile, traditional) = 0;
    }

  if (CPP_OPTION (pfile, warn_trigraphs) == 2)
    CPP_OPTION (pfile, warn_trigraphs) = !CPP_OPTION (pfile, trigraphs);

  if (CPP_OPTION (pfile, traditional))
    {
      /* Traditional CPP does not accurately track column information.  */
      CPP_OPTION (pfile, show_column) = 0;
      CPP_OPTION (pfile, trigraphs) = 0;
      CPP_OPTION (pfile, warn_trigraphs) = 0;
    }
}
2667'>2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387 4388 4389 4390 4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 4409 4410 4411 4412 4413 4414 4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460 4461 4462 4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510 4511 4512 4513 4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526 4527 4528 4529 4530 4531 4532 4533 4534 4535 4536 4537 4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548 4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715 4716 4717 4718 4719 4720 4721 4722 4723 4724 4725 4726 4727 4728 4729 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786 4787 4788 4789 4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814 4815 4816 4817 4818 4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860 4861 4862 4863 4864 4865 4866 4867 4868 4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896 4897 4898 4899 4900 4901 4902 4903 4904 4905 4906 4907 4908 4909 4910 4911 4912 4913 4914 4915 4916 4917 4918 4919 4920 4921 4922 4923 4924 4925 4926 4927 4928 4929 4930 4931 4932 4933 4934 4935 4936 4937 4938 4939 4940 4941 4942 4943 4944 4945 4946 4947 4948 4949 4950 4951 4952 4953 4954 4955 4956 4957 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970 4971 4972 4973 4974 4975 4976 4977 4978 4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990 4991 4992 4993 4994 4995 4996 4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015 5016 5017 5018 5019 5020 5021 5022 5023 5024 5025 5026 5027 5028 5029 5030 5031 5032 5033 5034 5035 5036 5037 5038 5039 5040 5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056 5057 5058 5059 5060 5061 5062 5063 5064 5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081 5082 5083 5084 5085 5086 5087 5088 5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099 5100 5101 5102 5103 5104 5105 5106 5107 5108 5109 5110 5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144 5145 5146 5147 5148 5149 5150 5151 5152 5153 5154 5155 5156 5157 5158 5159 5160 5161 5162 5163 5164 5165 5166 5167 5168 5169 5170 5171 5172 5173 5174 5175 5176 5177 5178 5179 5180 5181 5182 5183 5184 5185 5186 5187 5188 5189 5190 5191 5192 5193 5194 5195 5196 5197 5198 5199 5200 5201 5202 5203 5204 5205 5206 5207 5208 5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225 5226 5227 5228 5229 5230 5231 5232 5233 5234 5235 5236 5237 5238 5239 5240 5241 5242 5243 5244 5245 5246 5247 5248 5249 5250 5251 5252 5253 5254 5255 5256 5257 5258 5259 5260 5261 5262 5263 5264 5265 5266 5267 5268 5269 5270 5271 5272 5273 5274 5275 5276 5277 5278 5279 5280 5281 5282 5283 5284 5285 5286 5287 5288 5289 5290 5291 5292 5293 5294 5295 5296 5297 5298 5299 5300 5301 5302 5303 5304 5305 5306 5307 5308 5309 5310 5311 5312 5313 5314 5315 5316 5317 5318 5319 5320 5321 5322 5323 5324 5325 5326 5327 5328 5329 5330 5331 5332 5333 5334 5335 5336 5337 5338 5339 5340 5341 5342 5343 5344 5345 5346 5347 5348 5349 5350 5351 5352 5353 5354 5355 5356 5357 5358 5359 5360 5361 5362 5363 5364 5365 5366 5367 5368 5369 5370 5371 5372 5373 5374 5375 5376 5377 5378 5379 5380 5381 5382 5383 5384 5385 5386 5387 5388 5389 5390 5391 5392 5393 5394 5395 5396 5397 5398 5399 5400 5401 5402 5403 5404 5405 5406 5407 5408 5409 5410 5411 5412 5413 5414 5415 5416 5417 5418 5419 5420 5421 5422 5423 5424 5425 5426 5427 5428 5429 5430 5431 5432 5433 5434 5435 5436 5437 5438 5439 5440 5441 5442 5443 5444 5445 5446 5447 5448 5449 5450 5451 5452 5453 5454 5455 5456 5457 5458 5459 5460 5461 5462 5463 5464 5465 5466 5467 5468 5469 5470 5471 5472 5473 5474 5475 5476 5477 5478 5479 5480 5481 5482 5483 5484 5485 5486 5487 5488 5489 5490 5491 5492 5493 5494 5495 5496 5497 5498 5499 5500 5501 5502 5503 5504 5505 5506 5507 5508 5509 5510 5511 5512 5513 5514 5515 5516 5517 5518 5519 5520 5521 5522 5523 5524 5525 5526 5527 5528 5529 5530 5531 5532 5533 5534 5535 5536 5537 5538 5539 5540 5541 5542 5543 5544 5545 5546 5547 5548 5549 5550 5551 5552 5553 5554 5555 5556 5557 5558 5559 5560 5561 5562 5563 5564 5565 5566 5567 5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578 5579 5580 5581 5582 5583 5584 5585 5586 5587 5588 5589 5590 5591 5592 5593 5594 5595 5596 5597 5598 5599 5600 5601 5602 5603 5604 5605 5606 5607 5608 5609 5610 5611 5612 5613 5614 5615 5616 5617 5618 5619 5620 5621 5622 5623 5624 5625 5626 5627 5628 5629 5630 5631 5632 5633 5634 5635 5636 5637 5638 5639 5640 5641 5642 5643 5644 5645 5646 5647 5648 5649 5650 5651 5652 5653 5654 5655 5656 5657 5658 5659 5660 5661 5662 5663 5664 5665 5666 5667 5668 5669 5670 5671 5672 5673 5674 5675 5676 5677 5678 5679 5680 5681 5682 5683 5684 5685 5686 5687 5688 5689 5690 5691 5692 5693 5694 5695 5696 5697 5698 5699 5700 5701 5702 5703 5704 5705 5706 5707 5708 5709 5710 5711 5712 5713 5714 5715 5716 5717 5718 5719 5720 5721 5722 5723 5724 5725 5726 5727 5728 5729 5730 5731 5732 5733 5734 5735 5736 5737 5738 5739 5740 5741 5742 5743 5744 5745 5746 5747 5748 5749 5750 5751 5752 5753 5754 5755 5756 5757 5758 5759 5760 5761 5762 5763 5764 5765 5766 5767 5768 5769 5770 5771 5772 5773 5774 5775 5776 5777 5778 5779 5780 5781 5782 5783 5784 5785 5786 5787 5788 5789 5790 5791 5792 5793 5794 5795 5796 5797 5798 5799 5800 5801 5802 5803 5804 5805 5806 5807 5808 5809 5810 5811 5812 5813 5814 5815 5816 5817 5818 5819 5820 5821 5822 5823 5824 5825 5826 5827 5828 5829 5830 5831 5832 5833 5834 5835 5836 5837 5838 5839 5840 5841 5842 5843 5844 5845 5846 5847 5848 5849 5850 5851 5852 5853 5854 5855 5856 5857 5858 5859 5860 5861 5862 5863 5864 5865 5866 5867 5868 5869 5870 5871 5872 5873 5874 5875 5876 5877 5878 5879 5880 5881 5882 5883 5884 5885 5886 5887 5888 5889 5890 5891 5892 5893 5894 5895 5896 5897 5898 5899 5900 5901 5902 5903 5904 5905 5906 5907 5908 5909 5910 5911 5912 5913 5914 5915 5916 5917 5918 5919 5920 5921 5922 5923 5924 5925 5926 5927 5928 5929 5930 5931 5932 5933 5934 5935 5936 5937 5938 5939 5940 5941 5942 5943 5944 5945 5946 5947 5948 5949 5950 5951 5952 5953 5954 5955 5956 5957 5958 5959 5960 5961 5962 5963 5964 5965 5966 5967 5968 5969 5970 5971 5972 5973 5974 5975 5976 5977 5978 5979 5980 5981 5982 5983 5984 5985 5986 5987 5988 5989 5990 5991 5992 5993 5994 5995 5996 5997 5998 5999 6000 6001 6002 6003 6004 6005 6006 6007 6008 6009 6010 6011 6012 6013 6014 6015 6016 6017 6018 6019 6020 6021 6022 6023 6024 6025 6026 6027 6028 6029 6030 6031 6032 6033 6034 6035 6036 6037 6038 6039 6040 6041 6042 6043 6044 6045 6046 6047 6048 6049 6050 6051 6052 6053 6054 6055 6056 6057 6058 6059 6060 6061 6062 6063 6064 6065 6066 6067 6068 6069 6070 6071 6072 6073 6074 6075 6076 6077 6078 6079 6080 6081 6082 6083 6084 6085 6086 6087 6088 6089 6090 6091 6092 6093 6094 6095 6096 6097 6098 6099 6100 6101 6102 6103 6104 6105 6106 6107 6108 6109 6110 6111 6112 6113 6114 6115 6116 6117 6118 6119 6120 6121 6122 6123 6124 6125 6126 6127 6128 6129 6130 6131 6132 6133 6134 6135 6136 6137 6138 6139 6140 6141 6142 6143 6144 6145 6146 6147 6148 6149 6150 6151 6152 6153 6154 6155 6156 6157 6158 6159 6160 6161 6162 6163 6164 6165 6166 6167 6168 6169 6170 6171 6172 6173 6174 6175 6176 6177 6178 6179 6180 6181 6182 6183 6184 6185 6186 6187 6188 6189 6190 6191 6192 6193 6194 6195 6196 6197 6198 6199 6200 6201 6202 6203 6204 6205 6206 6207 6208 6209 6210 6211 6212 6213 6214 6215 6216 6217 6218 6219 6220 6221 6222 6223 6224 6225 6226 6227 6228 6229 6230 6231 6232 6233 6234 6235 6236 6237 6238 6239 6240 6241 6242 6243 6244 6245 6246 6247 6248 6249 6250 6251 6252 6253 6254 6255 6256 6257 6258 6259 6260 6261 6262 6263 6264 6265 6266 6267 6268 6269 6270 6271 6272 6273 6274 6275 6276 6277 6278 6279 6280 6281 6282 6283 6284 6285 6286 6287 6288 6289 6290 6291 6292 6293 6294 6295 6296 6297 6298 6299 6300 6301 6302 6303 6304 6305 6306 6307 6308 6309 6310 6311 6312 6313 6314 6315 6316 6317 6318 6319 6320 6321 6322 6323 6324 6325 6326 6327 6328 6329 6330 6331 6332 6333 6334 6335 6336 6337 6338 6339 6340 6341 6342 6343 6344 6345 6346 6347 6348 6349 6350 6351 6352 6353 6354 6355 6356 6357 6358 6359 6360 6361 6362 6363 6364 6365 6366 6367 6368 6369 6370 6371 6372 6373 6374 6375 6376 6377 6378 6379 6380 6381 6382 6383 6384 6385 6386 6387 6388 6389 6390 6391 6392 6393 6394 6395 6396 6397 6398 6399 6400 6401 6402 6403 6404 6405 6406 6407 6408 6409 6410 6411 6412 6413 6414 6415 6416 6417 6418 6419 6420 6421 6422 6423 6424 6425 6426 6427 6428 6429 6430 6431 6432 6433 6434 6435 6436 6437 6438 6439 6440 6441 6442 6443 6444 6445 6446 6447 6448 6449 6450 6451 6452 6453 6454 6455 6456 6457 6458 6459 6460 6461 6462 6463 6464 6465 6466 6467 6468 6469 6470 6471 6472 6473 6474 6475 6476 6477 6478 6479 6480 6481 6482 6483 6484 6485 6486 6487 6488 6489 6490 6491 6492 6493 6494 6495 6496 6497 6498 6499 6500 6501 6502 6503 6504 6505 6506 6507 6508 6509 6510 6511 6512 6513 6514 6515 6516 6517 6518 6519 6520 6521 6522 6523 6524 6525 6526 6527 6528 6529 6530 6531 6532 6533 6534 6535 6536 6537 6538 6539 6540 6541 6542 6543 6544 6545 6546 6547 6548 6549 6550 6551 6552 6553 6554 6555 6556 6557 6558 6559 6560 6561 6562 6563 6564 6565 6566 6567 6568 6569 6570 6571 6572 6573 6574 6575 6576 6577 6578 6579 6580 6581 6582 6583 6584 6585 6586 6587 6588 6589 6590 6591 6592 6593 6594 6595 6596 6597 6598 6599 6600 6601 6602 6603 6604 6605 6606 6607 6608 6609 6610 6611 6612 6613 6614 6615 6616 6617 6618 6619 6620 6621 6622 6623 6624 6625 6626 6627 6628 6629 6630 6631 6632 6633 6634 6635 6636 6637 6638 6639 6640 6641 6642 6643 6644 6645 6646 6647 6648 6649 6650 6651 6652 6653 6654 6655 6656 6657 6658 6659 6660 6661 6662 6663 6664 6665 6666 6667 6668 6669 6670 6671 6672 6673 6674 6675 6676 6677 6678 6679 6680 6681 6682 6683 6684 6685 6686 6687 6688 6689 6690 6691 6692 6693 6694 6695 6696 6697 6698 6699 6700 6701 6702 6703 6704 6705 6706 6707 6708 6709 6710 6711 6712 6713 6714 6715 6716 6717 6718 6719 6720 6721 6722 6723 6724 6725 6726 6727 6728 6729 6730 6731 6732 6733 6734 6735 6736 6737 6738 6739 6740 6741 6742 6743 6744 6745 6746 6747 6748 6749 6750 6751 6752 6753 6754 6755 6756 6757 6758 6759 6760 6761 6762 6763 6764 6765 6766 6767 6768 6769 6770 6771 6772 6773 6774 6775 6776 6777 6778 6779 6780 6781 6782 6783 6784 6785 6786 6787 6788 6789 6790 6791 6792 6793 6794 6795 6796 6797 6798 6799 6800 6801 6802 6803 6804 6805 6806 6807 6808 6809 6810 6811 6812 6813 6814 6815 6816 6817 6818 6819 6820 6821 6822 6823 6824 6825 6826 6827 6828 6829 6830 6831 6832 6833 6834 6835 6836 6837 6838 6839 6840 6841 6842 6843 6844 6845 6846 6847 6848 6849 6850 6851 6852 6853 6854 6855 6856 6857 6858 6859 6860 6861 6862 6863 6864 6865 6866 6867 6868 6869 6870 6871 6872 6873 6874 6875 6876 6877 6878 6879 6880 6881 6882 6883 6884 6885 6886 6887 6888 6889 6890 6891 6892 6893 6894 6895 6896 6897 6898 6899 6900 6901 6902 6903 6904 6905 6906 6907 6908 6909 6910 6911 6912 6913 6914 6915 6916 6917 6918 6919 6920 6921 6922 6923 6924 6925 6926 6927 6928 6929 6930 6931 6932 6933 6934 6935 6936 6937 6938 6939 6940 6941 6942 6943 6944 6945 6946 6947 6948 6949 6950 6951 6952 6953 6954 6955 6956 6957 6958 6959 6960 6961 6962 6963 6964 6965 6966 6967 6968 6969 6970 6971 6972 6973 6974 6975 6976 6977 6978 6979 6980 6981 6982 6983 6984 6985 6986 6987 6988 6989 6990 6991 6992 6993 6994 6995 6996 6997 6998 6999 7000 7001 7002 7003 7004 7005 7006 7007 7008 7009 7010 7011 7012 7013 7014 7015 7016 7017 7018 7019 7020 7021 7022 7023 7024 7025 7026 7027 7028 7029 7030 7031 7032 7033 7034 7035 7036 7037 7038 7039 7040 7041 7042 7043 7044 7045 7046 7047 7048 7049 7050 7051 7052 7053 7054 7055 7056 7057 7058 7059 7060 7061 7062 7063 7064 7065 7066 7067 7068 7069 7070 7071 7072 7073 7074 7075 7076 7077 7078 7079 7080 7081 7082 7083 7084 7085 7086 7087 7088 7089 7090 7091 7092 7093 7094 7095 7096 7097 7098 7099 7100 7101 7102 7103 7104 7105 7106 7107 7108 7109 7110 7111 7112 7113 7114 7115 7116 7117 7118 7119 7120 7121 7122 7123 7124 7125 7126 7127 7128 7129 7130 7131 7132 7133 7134 7135 7136 7137 7138 7139 7140 7141 7142 7143 7144 7145 7146 7147 7148 7149 7150 7151 7152 7153 7154 7155 7156 7157 7158 7159 7160 7161 7162 7163 7164 7165 7166 7167 7168 7169 7170 7171 7172 7173 7174 7175 7176 7177 7178 7179 7180 7181 7182 7183 7184 7185 7186 7187 7188 7189 7190 7191 7192 7193 7194 7195 7196 7197 7198 7199 7200 7201 7202 7203 7204 7205 7206 7207 7208 7209 7210 7211 7212 7213 7214 7215 7216 7217 7218 7219 7220 7221 7222 7223 7224 7225 7226 7227 7228 7229 7230 7231 7232 7233 7234 7235 7236 7237 7238 7239 7240 7241 7242 7243 7244 7245 7246 7247 7248 7249 7250 7251 7252 7253 7254 7255 7256 7257 7258 7259 7260 7261 7262 7263 7264 7265 7266 7267 7268 7269 7270 7271 7272 7273 7274 7275 7276 7277 7278 7279 7280 7281 7282 7283 7284 7285 7286 7287 7288 7289 7290 7291 7292 7293 7294 7295 7296 7297 7298 7299 7300 7301 7302 7303 7304 7305 7306 7307 7308 7309 7310 7311 7312 7313 7314 7315 7316 7317 7318 7319 7320 7321 7322 7323 7324 7325 7326 7327 7328 7329 7330 7331 7332 7333 7334 7335 7336 7337 7338 7339 7340 7341 7342 7343 7344 7345 7346 7347 7348 7349 7350 7351 7352 7353 7354 7355 7356 7357 7358 7359 7360 7361 7362 7363 7364 7365 7366 7367 7368 7369 7370 7371 7372 7373 7374 7375 7376 7377 7378 7379 7380 7381 7382 7383 7384 7385 7386 7387 7388 7389 7390 7391 7392 7393 7394 7395 7396 7397 7398 7399 7400 7401 7402 7403 7404 7405 7406 7407 7408 7409 7410 7411 7412 7413 7414 7415 7416 7417 7418 7419 7420 7421 7422 7423 7424 7425 7426 7427 7428 7429 7430 7431 7432 7433 7434 7435 7436 7437 7438 7439 7440 7441 7442 7443 7444 7445 7446 7447 7448 7449 7450 7451 7452 7453 7454 7455 7456 7457 7458 7459 7460 7461 7462 7463 7464 7465 7466 7467 7468 7469 7470 7471 7472 7473 7474 7475 7476 7477 7478 7479 7480 7481 7482 7483 7484 7485 7486 7487 7488 7489 7490 7491 7492 7493 7494 7495 7496 7497 7498 7499 7500 7501 7502 7503 7504 7505 7506 7507 7508 7509 7510 7511 7512 7513 7514 7515 7516 7517 7518 7519 7520 7521 7522 7523 7524 7525 7526 7527 7528 7529 7530 7531 7532 7533 7534 7535 7536 7537 7538 7539 7540 7541 7542 7543 7544 7545 7546 7547 7548 7549 7550 7551 7552 7553 7554 7555 7556 7557 7558 7559 7560 7561 7562 7563 7564 7565 7566 7567 7568 7569 7570 7571 7572 7573 7574 7575 7576 7577 7578 7579 7580 7581 7582 7583 7584 7585 7586 7587 7588 7589 7590 7591 7592 7593 7594 7595 7596 7597 7598 7599 7600 7601 7602 7603 7604 7605 7606 7607 7608 7609 7610 7611 7612 7613 7614 7615 7616 7617 7618 7619 7620 7621 7622 7623 7624 7625 7626 7627 7628 7629 7630 7631 7632 7633 7634 7635 7636 7637 7638 7639 7640 7641 7642 7643 7644 7645 7646 7647 7648 7649 7650 7651 7652 7653 7654 7655 7656 7657 7658 7659 7660 7661 7662 7663 7664 7665 7666 7667 7668 7669 7670 7671 7672 7673 7674 7675 7676 7677 7678 7679 7680 7681 7682 7683 7684 7685 7686 7687 7688 7689 7690 7691 7692 7693 7694 7695 7696 7697 7698 7699 7700 7701 7702 7703 7704 7705 7706 7707 7708 7709 7710 7711 7712 7713 7714 7715 7716 7717 7718 7719 7720 7721 7722 7723 7724 7725 7726 7727 7728 7729 7730 7731 7732 7733 7734 7735 7736 7737 7738 7739 7740 7741 7742 7743 7744 7745 7746 7747 7748 7749 7750 7751 7752 7753 7754 7755 7756 7757 7758 7759 7760 7761 7762 7763 7764 7765 7766 7767 7768 7769 7770 7771 7772 7773 7774 7775 7776 7777 7778 7779 7780 7781 7782 7783 7784 7785 7786 7787 7788 7789 7790 7791 7792 7793 7794 7795 7796 7797 7798 7799 7800 7801 7802 7803 7804 7805 7806 7807 7808 7809 7810 7811 7812 7813 7814 7815 7816 7817 7818 7819 7820 7821 7822 7823 7824 7825 7826 7827 7828 7829 7830 7831 7832 7833 7834 7835 7836 7837 7838 7839 7840 7841 7842 7843 7844 7845 7846 7847 7848 7849 7850 7851 7852 7853 7854 7855 7856 7857 7858 7859 7860 7861 7862 7863 7864 7865 7866 7867 7868 7869 7870 7871 7872 7873 7874 7875 7876 7877 7878 7879 7880 7881 7882 7883 7884 7885 7886 7887 7888 7889 7890 7891 7892 7893 7894 7895 7896 7897 7898 7899 7900 7901 7902 7903 7904 7905 7906 7907 7908 7909 7910 7911 7912 7913 7914 7915 7916 7917 7918 7919 7920 7921 7922 7923 7924 7925 7926 7927 7928 7929 7930 7931 7932 7933 7934 7935 7936 7937 7938 7939 7940 7941 7942 7943 7944 7945 7946 7947 7948 7949 7950 7951 7952 7953 7954 7955 7956 7957 7958 7959 7960 7961 7962 7963 7964 7965 7966 7967 7968 7969 7970 7971 7972 7973 7974 7975 7976 7977 7978 7979 7980 7981 7982 7983 7984 7985 7986 7987 7988 7989 7990 7991 7992 7993 7994 7995 7996 7997 7998 7999 8000 8001 8002 8003 8004 8005 8006 8007 8008 8009 8010 8011 8012 8013 8014 8015 8016 8017 8018 8019 8020 8021 8022 8023 8024 8025 8026 8027 8028 8029 8030 8031 8032 8033 8034 8035 8036 8037 8038 8039 8040 8041 8042 8043 8044 8045 8046 8047 8048 8049 8050 8051 8052 8053 8054 8055 8056 8057 8058 8059 8060 8061 8062 8063 8064 8065 8066 8067 8068 8069 8070 8071 8072 8073 8074 8075 8076 8077 8078 8079 8080 8081 8082 8083 8084 8085 8086 8087 8088 8089 8090 8091 8092 8093 8094 8095 8096 8097 8098 8099 8100 8101 8102 8103 8104 8105 8106 8107 8108 8109 8110 8111 8112 8113 8114 8115 8116 8117 8118 8119 8120 8121 8122 8123 8124 8125 8126 8127 8128 8129 8130 8131 8132 8133 8134 8135 8136 8137 8138 8139 8140 8141 8142 8143 8144 8145 8146 8147 8148 8149 8150 8151 8152 8153 8154 8155 8156 8157 8158 8159 8160 8161 8162 8163 8164 8165 8166 8167 8168 8169 8170 8171 8172 8173 8174 8175 8176 8177 8178 8179 8180 8181 8182 8183 8184 8185 8186 8187 8188 8189 8190 8191 8192 8193 8194 8195 8196 8197 8198 8199 8200 8201 8202 8203 8204 8205 8206 8207 8208 8209 8210 8211 8212 8213 8214 8215 8216 8217 8218 8219 8220 8221 8222 8223 8224 8225 8226 8227 8228 8229 8230 8231 8232 8233 8234 8235 8236 8237 8238 8239 8240 8241 8242 8243 8244 8245 8246 8247 8248 8249 8250 8251 8252 8253 8254 8255 8256 8257 8258 8259 8260 8261 8262 8263 8264 8265 8266 8267 8268 8269 8270 8271 8272 8273 8274 8275 8276 8277 8278 8279 8280 8281 8282 8283 8284 8285 8286 8287 8288 8289 8290 8291 8292 8293 8294 8295 8296 8297 8298 8299 8300 8301 8302 8303 8304 8305 8306 8307 8308 8309 8310 8311 8312 8313 8314 8315 8316 8317 8318 8319 8320 8321 8322 8323 8324 8325 8326 8327 8328 8329 8330 8331 8332 8333 8334 8335 8336 8337 8338 8339 8340 8341 8342 8343 8344 8345 8346 8347 8348 8349 8350 8351 8352 8353 8354 8355 8356 8357 8358 8359 8360 8361 8362 8363 8364 8365 8366 8367 8368 8369 8370 8371 8372 8373 8374 8375 8376 8377 8378 8379 8380 8381 8382 8383 8384 8385 8386 8387 8388 8389 8390 8391 8392 8393 8394 8395 8396 8397 8398 8399 8400 8401 8402 8403 8404 8405 8406 8407 8408 8409 8410 8411 8412 8413 8414 8415 8416 8417 8418 8419 8420 8421 8422 8423 8424 8425 8426 8427 8428 8429 8430 8431 8432 8433 8434 8435 8436 8437 8438 8439 8440 8441 8442 8443 8444 8445 8446 8447 8448 8449 8450 8451 8452 8453 8454 8455 8456 8457 8458 8459 8460 8461 8462 8463 8464 8465 8466 8467 8468 8469 8470 8471 8472 8473 8474 8475 8476 8477 8478 8479 8480 8481 8482 8483 8484 8485 8486 8487 8488 8489 8490 8491 8492 8493 8494 8495 8496 8497 8498 8499 8500 8501 8502 8503 8504 8505 8506 8507 8508 8509 8510 8511 8512 8513 8514 8515 8516 8517 8518 8519 8520 8521 8522 8523 8524 8525 8526 8527 8528 8529 8530 8531 8532 8533 8534 8535 8536 8537 8538 8539 8540 8541 8542 8543 8544 8545 8546 8547 8548 8549 8550 8551 8552 8553 8554 8555 8556 8557 8558 8559 8560 8561 8562 8563 8564 8565 8566 8567 8568 8569 8570 8571 8572 8573 8574 8575 8576 8577 8578 8579 8580 8581 8582 8583 8584 8585 8586 8587 8588 8589 8590 8591 8592 8593 8594 8595 8596 8597 8598 8599 8600 8601 8602 8603 8604 8605 8606 8607 8608 8609 8610 8611 8612 8613 8614 8615 8616 8617 8618 8619 8620 8621 8622 8623 8624 8625 8626 8627 8628 8629 8630 8631 8632 8633 8634 8635 8636 8637 8638 8639 8640 8641 8642 8643 8644 8645 8646 8647 8648 8649 8650 8651 8652 8653 8654 8655 8656 8657 8658 8659 8660 8661 8662 8663 8664 8665 8666 8667 8668 8669 8670 8671 8672 8673 8674 8675 8676 8677 8678 8679 8680 8681 8682 8683 8684 8685 8686 8687 8688 8689 8690 8691 8692 8693 8694 8695 8696 8697 8698 8699 8700 8701 8702 8703 8704 8705 8706 8707 8708 8709 8710 8711 8712 8713 8714 8715 8716 8717 8718 8719 8720 8721 8722 8723 8724 8725 8726 8727 8728 8729 8730 8731 8732 8733 8734 8735 8736 8737 8738 8739 8740 8741 8742 8743 8744 8745 8746 8747 8748 8749 8750 8751 8752 8753 8754 8755 8756 8757 8758 8759 8760 8761 8762 8763 8764 8765 8766 8767 8768 8769 8770 8771 8772 8773 8774 8775 8776 8777 8778 8779 8780 8781 8782 8783 8784 8785 8786 8787 8788 8789 8790 8791 8792 8793 8794 8795 8796 8797 8798 8799 8800 8801 8802 8803 8804 8805 8806 8807 8808 8809 8810 8811 8812 8813 8814 8815 8816 8817 8818 8819 8820 8821 8822 8823 8824 8825 8826 8827 8828 8829 8830 8831 8832 8833 8834 8835 8836 8837 8838 8839 8840 8841 8842 8843 8844 8845 8846 8847 8848 8849 8850 8851 8852 8853 8854 8855 8856 8857 8858 8859 8860 8861 8862 8863 8864 8865 8866 8867 8868 8869 8870 8871 8872 8873 8874 8875 8876 8877 8878 8879 8880 8881 8882 8883 8884 8885 8886 8887 8888 8889 8890 8891 8892 8893 8894 8895 8896 8897 8898 8899 8900 8901 8902 8903 8904 8905 8906 8907 8908 8909 8910 8911 8912 8913 8914 8915 8916 8917 8918 8919 8920 8921 8922 8923 8924 8925 8926 8927 8928 8929 8930 8931 8932 8933 8934 8935 8936 8937 8938 8939 8940 8941 8942 8943 8944 8945 8946 8947 8948 8949 8950 8951 8952 8953 8954 8955 8956 8957 8958 8959 8960 8961 8962 8963 8964 8965 8966 8967 8968 8969 8970 8971 8972 8973 8974 8975 8976 8977 8978 8979 8980 8981 8982 8983 8984 8985 8986 8987 8988 8989 8990 8991 8992 8993 8994 8995 8996 8997 8998 8999 9000 9001 9002 9003 9004 9005 9006 9007 9008 9009 9010 9011 9012 9013 9014 9015 9016 9017 9018 9019 9020 9021 9022 9023 9024 9025 9026 9027 9028 9029 9030 9031 9032 9033 9034 9035 9036 9037 9038 9039 9040 9041 9042 9043 9044 9045 9046 9047 9048 9049 9050 9051 9052 9053 9054 9055 9056 9057 9058 9059 9060 9061 9062 9063 9064 9065 9066 9067 9068 9069 9070 9071 9072 9073 9074 9075 9076 9077 9078 9079 9080 9081 9082 9083 9084 9085 9086 9087 9088 9089 9090 9091 9092 9093 9094 9095 9096 9097 9098 9099 9100 9101 9102 9103 9104 9105 9106 9107 9108 9109 9110 9111 9112 9113 9114 9115 9116 9117 9118 9119 9120 9121 9122 9123 9124 9125 9126 9127 9128 9129 9130 9131 9132 9133 9134 9135 9136 9137 9138 9139 9140 9141 9142 9143 9144 9145 9146 9147 9148 9149 9150 9151 9152 9153 9154 9155 9156 9157 9158 9159 9160 9161 9162 9163 9164 9165 9166 9167 9168 9169 9170 9171 9172 9173 9174 9175 9176 9177 9178 9179 9180 9181 9182 9183 9184 9185 9186 9187 9188 9189 9190 9191 9192 9193 9194 9195 9196 9197 9198 9199 9200 9201 9202 9203 9204 9205 9206 9207 9208 9209 9210 9211 9212 9213 9214 9215 9216 9217 9218 9219 9220 9221 9222 9223 9224 9225 9226 9227 9228 9229 9230 9231 9232 9233 9234 9235 9236 9237 9238 9239 9240 9241 9242
/* Loop Vectorization
   Copyright (C) 2003-2020 Free Software Foundation, Inc.
   Contributed by Dorit Naishlos <dorit@il.ibm.com> and
   Ira Rosen <irar@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/>.  */

#include "config.h"
#include "system.h"
#include "coretypes.h"
#include "backend.h"
#include "target.h"
#include "rtl.h"
#include "tree.h"
#include "gimple.h"
#include "cfghooks.h"
#include "tree-pass.h"
#include "ssa.h"
#include "optabs-tree.h"
#include "diagnostic-core.h"
#include "fold-const.h"
#include "stor-layout.h"
#include "cfganal.h"
#include "gimplify.h"
#include "gimple-iterator.h"
#include "gimplify-me.h"
#include "tree-ssa-loop-ivopts.h"
#include "tree-ssa-loop-manip.h"
#include "tree-ssa-loop-niter.h"
#include "tree-ssa-loop.h"
#include "cfgloop.h"
#include "tree-scalar-evolution.h"
#include "tree-vectorizer.h"
#include "gimple-fold.h"
#include "cgraph.h"
#include "tree-cfg.h"
#include "tree-if-conv.h"
#include "internal-fn.h"
#include "tree-vector-builder.h"
#include "vec-perm-indices.h"
#include "tree-eh.h"

/* Loop Vectorization Pass.

   This pass tries to vectorize loops.

   For example, the vectorizer transforms the following simple loop:

        short a[N]; short b[N]; short c[N]; int i;

        for (i=0; i<N; i++){
          a[i] = b[i] + c[i];
        }

   as if it was manually vectorized by rewriting the source code into:

        typedef int __attribute__((mode(V8HI))) v8hi;
        short a[N];  short b[N]; short c[N];   int i;
        v8hi *pa = (v8hi*)a, *pb = (v8hi*)b, *pc = (v8hi*)c;
        v8hi va, vb, vc;

        for (i=0; i<N/8; i++){
          vb = pb[i];
          vc = pc[i];
          va = vb + vc;
          pa[i] = va;
        }

        The main entry to this pass is vectorize_loops(), in which
   the vectorizer applies a set of analyses on a given set of loops,
   followed by the actual vectorization transformation for the loops that
   had successfully passed the analysis phase.
        Throughout this pass we make a distinction between two types of
   data: scalars (which are represented by SSA_NAMES), and memory references
   ("data-refs").  These two types of data require different handling both
   during analysis and transformation. The types of data-refs that the
   vectorizer currently supports are ARRAY_REFS which base is an array DECL
   (not a pointer), and INDIRECT_REFS through pointers; both array and pointer
   accesses are required to have a simple (consecutive) access pattern.

   Analysis phase:
   ===============
        The driver for the analysis phase is vect_analyze_loop().
   It applies a set of analyses, some of which rely on the scalar evolution
   analyzer (scev) developed by Sebastian Pop.

        During the analysis phase the vectorizer records some information
   per stmt in a "stmt_vec_info" struct which is attached to each stmt in the
   loop, as well as general information about the loop as a whole, which is
   recorded in a "loop_vec_info" struct attached to each loop.

   Transformation phase:
   =====================
        The loop transformation phase scans all the stmts in the loop, and
   creates a vector stmt (or a sequence of stmts) for each scalar stmt S in
   the loop that needs to be vectorized.  It inserts the vector code sequence
   just before the scalar stmt S, and records a pointer to the vector code
   in STMT_VINFO_VEC_STMT (stmt_info) (stmt_info is the stmt_vec_info struct
   attached to S).  This pointer will be used for the vectorization of following
   stmts which use the def of stmt S. Stmt S is removed if it writes to memory;
   otherwise, we rely on dead code elimination for removing it.

        For example, say stmt S1 was vectorized into stmt VS1:

   VS1: vb = px[i];
   S1:  b = x[i];    STMT_VINFO_VEC_STMT (stmt_info (S1)) = VS1
   S2:  a = b;

   To vectorize stmt S2, the vectorizer first finds the stmt that defines
   the operand 'b' (S1), and gets the relevant vector def 'vb' from the
   vector stmt VS1 pointed to by STMT_VINFO_VEC_STMT (stmt_info (S1)).  The
   resulting sequence would be:

   VS1: vb = px[i];
   S1:  b = x[i];       STMT_VINFO_VEC_STMT (stmt_info (S1)) = VS1
   VS2: va = vb;
   S2:  a = b;          STMT_VINFO_VEC_STMT (stmt_info (S2)) = VS2

        Operands that are not SSA_NAMEs, are data-refs that appear in
   load/store operations (like 'x[i]' in S1), and are handled differently.

   Target modeling:
   =================
        Currently the only target specific information that is used is the
   size of the vector (in bytes) - "TARGET_VECTORIZE_UNITS_PER_SIMD_WORD".
   Targets that can support different sizes of vectors, for now will need
   to specify one value for "TARGET_VECTORIZE_UNITS_PER_SIMD_WORD".  More
   flexibility will be added in the future.

        Since we only vectorize operations which vector form can be
   expressed using existing tree codes, to verify that an operation is
   supported, the vectorizer checks the relevant optab at the relevant
   machine_mode (e.g, optab_handler (add_optab, V8HImode)).  If
   the value found is CODE_FOR_nothing, then there's no target support, and
   we can't vectorize the stmt.

   For additional information on this project see:
   http://gcc.gnu.org/projects/tree-ssa/vectorization.html
*/

static void vect_estimate_min_profitable_iters (loop_vec_info, int *, int *);
static stmt_vec_info vect_is_simple_reduction (loop_vec_info, stmt_vec_info,
					       bool *, bool *);

/* Subroutine of vect_determine_vf_for_stmt that handles only one
   statement.  VECTYPE_MAYBE_SET_P is true if STMT_VINFO_VECTYPE
   may already be set for general statements (not just data refs).  */

static opt_result
vect_determine_vf_for_stmt_1 (vec_info *vinfo, stmt_vec_info stmt_info,
			      bool vectype_maybe_set_p,
			      poly_uint64 *vf)
{
  gimple *stmt = stmt_info->stmt;

  if ((!STMT_VINFO_RELEVANT_P (stmt_info)
       && !STMT_VINFO_LIVE_P (stmt_info))
      || gimple_clobber_p (stmt))
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location, "skip.\n");
      return opt_result::success ();
    }

  tree stmt_vectype, nunits_vectype;
  opt_result res = vect_get_vector_types_for_stmt (vinfo, stmt_info,
						   &stmt_vectype,
						   &nunits_vectype);
  if (!res)
    return res;

  if (stmt_vectype)
    {
      if (STMT_VINFO_VECTYPE (stmt_info))
	/* The only case when a vectype had been already set is for stmts
	   that contain a data ref, or for "pattern-stmts" (stmts generated
	   by the vectorizer to represent/replace a certain idiom).  */
	gcc_assert ((STMT_VINFO_DATA_REF (stmt_info)
		     || vectype_maybe_set_p)
		    && STMT_VINFO_VECTYPE (stmt_info) == stmt_vectype);
      else
	STMT_VINFO_VECTYPE (stmt_info) = stmt_vectype;
    }

  if (nunits_vectype)
    vect_update_max_nunits (vf, nunits_vectype);

  return opt_result::success ();
}

/* Subroutine of vect_determine_vectorization_factor.  Set the vector
   types of STMT_INFO and all attached pattern statements and update
   the vectorization factor VF accordingly.  Return true on success
   or false if something prevented vectorization.  */

static opt_result
vect_determine_vf_for_stmt (vec_info *vinfo,
			    stmt_vec_info stmt_info, poly_uint64 *vf)
{
  if (dump_enabled_p ())
    dump_printf_loc (MSG_NOTE, vect_location, "==> examining statement: %G",
		     stmt_info->stmt);
  opt_result res = vect_determine_vf_for_stmt_1 (vinfo, stmt_info, false, vf);
  if (!res)
    return res;

  if (STMT_VINFO_IN_PATTERN_P (stmt_info)
      && STMT_VINFO_RELATED_STMT (stmt_info))
    {
      gimple *pattern_def_seq = STMT_VINFO_PATTERN_DEF_SEQ (stmt_info);
      stmt_info = STMT_VINFO_RELATED_STMT (stmt_info);

      /* If a pattern statement has def stmts, analyze them too.  */
      for (gimple_stmt_iterator si = gsi_start (pattern_def_seq);
	   !gsi_end_p (si); gsi_next (&si))
	{
	  stmt_vec_info def_stmt_info = vinfo->lookup_stmt (gsi_stmt (si));
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_NOTE, vect_location,
			     "==> examining pattern def stmt: %G",
			     def_stmt_info->stmt);
	  res = vect_determine_vf_for_stmt_1 (vinfo, def_stmt_info, true, vf);
	  if (!res)
	    return res;
	}

      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location,
			 "==> examining pattern statement: %G",
			 stmt_info->stmt);
      res = vect_determine_vf_for_stmt_1 (vinfo, stmt_info, true, vf);
      if (!res)
	return res;
    }

  return opt_result::success ();
}

/* Function vect_determine_vectorization_factor

   Determine the vectorization factor (VF).  VF is the number of data elements
   that are operated upon in parallel in a single iteration of the vectorized
   loop.  For example, when vectorizing a loop that operates on 4byte elements,
   on a target with vector size (VS) 16byte, the VF is set to 4, since 4
   elements can fit in a single vector register.

   We currently support vectorization of loops in which all types operated upon
   are of the same size.  Therefore this function currently sets VF according to
   the size of the types operated upon, and fails if there are multiple sizes
   in the loop.

   VF is also the factor by which the loop iterations are strip-mined, e.g.:
   original loop:
        for (i=0; i<N; i++){
          a[i] = b[i] + c[i];
        }

   vectorized loop:
        for (i=0; i<N; i+=VF){
          a[i:VF] = b[i:VF] + c[i:VF];
        }
*/

static opt_result
vect_determine_vectorization_factor (loop_vec_info loop_vinfo)
{
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo);
  basic_block *bbs = LOOP_VINFO_BBS (loop_vinfo);
  unsigned nbbs = loop->num_nodes;
  poly_uint64 vectorization_factor = 1;
  tree scalar_type = NULL_TREE;
  gphi *phi;
  tree vectype;
  stmt_vec_info stmt_info;
  unsigned i;

  DUMP_VECT_SCOPE ("vect_determine_vectorization_factor");

  for (i = 0; i < nbbs; i++)
    {
      basic_block bb = bbs[i];

      for (gphi_iterator si = gsi_start_phis (bb); !gsi_end_p (si);
	   gsi_next (&si))
	{
	  phi = si.phi ();
	  stmt_info = loop_vinfo->lookup_stmt (phi);
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_NOTE, vect_location, "==> examining phi: %G",
			     phi);

	  gcc_assert (stmt_info);

	  if (STMT_VINFO_RELEVANT_P (stmt_info)
	      || STMT_VINFO_LIVE_P (stmt_info))
            {
	      gcc_assert (!STMT_VINFO_VECTYPE (stmt_info));
              scalar_type = TREE_TYPE (PHI_RESULT (phi));

	      if (dump_enabled_p ())
		dump_printf_loc (MSG_NOTE, vect_location,
				 "get vectype for scalar type:  %T\n",
				 scalar_type);

	      vectype = get_vectype_for_scalar_type (loop_vinfo, scalar_type);
	      if (!vectype)
		return opt_result::failure_at (phi,
					       "not vectorized: unsupported "
					       "data-type %T\n",
					       scalar_type);
	      STMT_VINFO_VECTYPE (stmt_info) = vectype;

	      if (dump_enabled_p ())
		dump_printf_loc (MSG_NOTE, vect_location, "vectype: %T\n",
				 vectype);

	      if (dump_enabled_p ())
		{
		  dump_printf_loc (MSG_NOTE, vect_location, "nunits = ");
		  dump_dec (MSG_NOTE, TYPE_VECTOR_SUBPARTS (vectype));
		  dump_printf (MSG_NOTE, "\n");
		}

	      vect_update_max_nunits (&vectorization_factor, vectype);
	    }
	}

      for (gimple_stmt_iterator si = gsi_start_bb (bb); !gsi_end_p (si);
	   gsi_next (&si))
	{
	  stmt_info = loop_vinfo->lookup_stmt (gsi_stmt (si));
	  opt_result res
	    = vect_determine_vf_for_stmt (loop_vinfo,
					  stmt_info, &vectorization_factor);
	  if (!res)
	    return res;
        }
    }

  /* TODO: Analyze cost. Decide if worth while to vectorize.  */
  if (dump_enabled_p ())
    {
      dump_printf_loc (MSG_NOTE, vect_location, "vectorization factor = ");
      dump_dec (MSG_NOTE, vectorization_factor);
      dump_printf (MSG_NOTE, "\n");
    }

  if (known_le (vectorization_factor, 1U))
    return opt_result::failure_at (vect_location,
				   "not vectorized: unsupported data-type\n");
  LOOP_VINFO_VECT_FACTOR (loop_vinfo) = vectorization_factor;
  return opt_result::success ();
}


/* Function vect_is_simple_iv_evolution.

   FORNOW: A simple evolution of an induction variables in the loop is
   considered a polynomial evolution.  */

static bool
vect_is_simple_iv_evolution (unsigned loop_nb, tree access_fn, tree * init,
                             tree * step)
{
  tree init_expr;
  tree step_expr;
  tree evolution_part = evolution_part_in_loop_num (access_fn, loop_nb);
  basic_block bb;

  /* When there is no evolution in this loop, the evolution function
     is not "simple".  */
  if (evolution_part == NULL_TREE)
    return false;

  /* When the evolution is a polynomial of degree >= 2
     the evolution function is not "simple".  */
  if (tree_is_chrec (evolution_part))
    return false;

  step_expr = evolution_part;
  init_expr = unshare_expr (initial_condition_in_loop_num (access_fn, loop_nb));

  if (dump_enabled_p ())
    dump_printf_loc (MSG_NOTE, vect_location, "step: %T,  init: %T\n",
		     step_expr, init_expr);

  *init = init_expr;
  *step = step_expr;

  if (TREE_CODE (step_expr) != INTEGER_CST
      && (TREE_CODE (step_expr) != SSA_NAME
	  || ((bb = gimple_bb (SSA_NAME_DEF_STMT (step_expr)))
	      && flow_bb_inside_loop_p (get_loop (cfun, loop_nb), bb))
	  || (!INTEGRAL_TYPE_P (TREE_TYPE (step_expr))
	      && (!SCALAR_FLOAT_TYPE_P (TREE_TYPE (step_expr))
		  || !flag_associative_math)))
      && (TREE_CODE (step_expr) != REAL_CST
	  || !flag_associative_math))
    {
      if (dump_enabled_p ())
        dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
                         "step unknown.\n");
      return false;
    }

  return true;
}

/* Return true if PHI, described by STMT_INFO, is the inner PHI in
   what we are assuming is a double reduction.  For example, given
   a structure like this:

      outer1:
	x_1 = PHI <x_4(outer2), ...>;
	...

      inner:
	x_2 = PHI <x_1(outer1), ...>;
	...
	x_3 = ...;
	...

      outer2:
	x_4 = PHI <x_3(inner)>;
	...

   outer loop analysis would treat x_1 as a double reduction phi and
   this function would then return true for x_2.  */

static bool
vect_inner_phi_in_double_reduction_p (loop_vec_info loop_vinfo, gphi *phi)
{
  use_operand_p use_p;
  ssa_op_iter op_iter;
  FOR_EACH_PHI_ARG (use_p, phi, op_iter, SSA_OP_USE)
    if (stmt_vec_info def_info = loop_vinfo->lookup_def (USE_FROM_PTR (use_p)))
      if (STMT_VINFO_DEF_TYPE (def_info) == vect_double_reduction_def)
	return true;
  return false;
}

/* Function vect_analyze_scalar_cycles_1.

   Examine the cross iteration def-use cycles of scalar variables
   in LOOP.  LOOP_VINFO represents the loop that is now being
   considered for vectorization (can be LOOP, or an outer-loop
   enclosing LOOP).  */

static void
vect_analyze_scalar_cycles_1 (loop_vec_info loop_vinfo, class loop *loop)
{
  basic_block bb = loop->header;
  tree init, step;
  auto_vec<stmt_vec_info, 64> worklist;
  gphi_iterator gsi;
  bool double_reduc, reduc_chain;

  DUMP_VECT_SCOPE ("vect_analyze_scalar_cycles");

  /* First - identify all inductions.  Reduction detection assumes that all the
     inductions have been identified, therefore, this order must not be
     changed.  */
  for (gsi = gsi_start_phis  (bb); !gsi_end_p (gsi); gsi_next (&gsi))
    {
      gphi *phi = gsi.phi ();
      tree access_fn = NULL;
      tree def = PHI_RESULT (phi);
      stmt_vec_info stmt_vinfo = loop_vinfo->lookup_stmt (phi);

      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location, "Analyze phi: %G", phi);

      /* Skip virtual phi's.  The data dependences that are associated with
         virtual defs/uses (i.e., memory accesses) are analyzed elsewhere.  */
      if (virtual_operand_p (def))
	continue;

      STMT_VINFO_DEF_TYPE (stmt_vinfo) = vect_unknown_def_type;

      /* Analyze the evolution function.  */
      access_fn = analyze_scalar_evolution (loop, def);
      if (access_fn)
	{
	  STRIP_NOPS (access_fn);
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_NOTE, vect_location,
			     "Access function of PHI: %T\n", access_fn);
	  STMT_VINFO_LOOP_PHI_EVOLUTION_BASE_UNCHANGED (stmt_vinfo)
	    = initial_condition_in_loop_num (access_fn, loop->num);
	  STMT_VINFO_LOOP_PHI_EVOLUTION_PART (stmt_vinfo)
	    = evolution_part_in_loop_num (access_fn, loop->num);
	}

      if (!access_fn
	  || vect_inner_phi_in_double_reduction_p (loop_vinfo, phi)
	  || !vect_is_simple_iv_evolution (loop->num, access_fn, &init, &step)
	  || (LOOP_VINFO_LOOP (loop_vinfo) != loop
	      && TREE_CODE (step) != INTEGER_CST))
	{
	  worklist.safe_push (stmt_vinfo);
	  continue;
	}

      gcc_assert (STMT_VINFO_LOOP_PHI_EVOLUTION_BASE_UNCHANGED (stmt_vinfo)
		  != NULL_TREE);
      gcc_assert (STMT_VINFO_LOOP_PHI_EVOLUTION_PART (stmt_vinfo) != NULL_TREE);

      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location, "Detected induction.\n");
      STMT_VINFO_DEF_TYPE (stmt_vinfo) = vect_induction_def;
    }


  /* Second - identify all reductions and nested cycles.  */
  while (worklist.length () > 0)
    {
      stmt_vec_info stmt_vinfo = worklist.pop ();
      gphi *phi = as_a <gphi *> (stmt_vinfo->stmt);
      tree def = PHI_RESULT (phi);

      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location, "Analyze phi: %G", phi);

      gcc_assert (!virtual_operand_p (def)
		  && STMT_VINFO_DEF_TYPE (stmt_vinfo) == vect_unknown_def_type);

      stmt_vec_info reduc_stmt_info
	= vect_is_simple_reduction (loop_vinfo, stmt_vinfo, &double_reduc,
				    &reduc_chain);
      if (reduc_stmt_info)
        {
	  STMT_VINFO_REDUC_DEF (stmt_vinfo) = reduc_stmt_info;
	  STMT_VINFO_REDUC_DEF (reduc_stmt_info) = stmt_vinfo;
	  if (double_reduc)
	    {
	      if (dump_enabled_p ())
		dump_printf_loc (MSG_NOTE, vect_location,
				 "Detected double reduction.\n");

              STMT_VINFO_DEF_TYPE (stmt_vinfo) = vect_double_reduction_def;
	      STMT_VINFO_DEF_TYPE (reduc_stmt_info) = vect_double_reduction_def;
            }
          else
            {
              if (loop != LOOP_VINFO_LOOP (loop_vinfo))
                {
                  if (dump_enabled_p ())
                    dump_printf_loc (MSG_NOTE, vect_location,
				     "Detected vectorizable nested cycle.\n");

                  STMT_VINFO_DEF_TYPE (stmt_vinfo) = vect_nested_cycle;
                }
              else
                {
                  if (dump_enabled_p ())
                    dump_printf_loc (MSG_NOTE, vect_location,
				     "Detected reduction.\n");

                  STMT_VINFO_DEF_TYPE (stmt_vinfo) = vect_reduction_def;
		  STMT_VINFO_DEF_TYPE (reduc_stmt_info) = vect_reduction_def;
                  /* Store the reduction cycles for possible vectorization in
                     loop-aware SLP if it was not detected as reduction
		     chain.  */
		  if (! reduc_chain)
		    LOOP_VINFO_REDUCTIONS (loop_vinfo).safe_push
		      (reduc_stmt_info);
                }
            }
        }
      else
        if (dump_enabled_p ())
          dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			   "Unknown def-use cycle pattern.\n");
    }
}


/* Function vect_analyze_scalar_cycles.

   Examine the cross iteration def-use cycles of scalar variables, by
   analyzing the loop-header PHIs of scalar variables.  Classify each
   cycle as one of the following: invariant, induction, reduction, unknown.
   We do that for the loop represented by LOOP_VINFO, and also to its
   inner-loop, if exists.
   Examples for scalar cycles:

   Example1: reduction:

              loop1:
              for (i=0; i<N; i++)
                 sum += a[i];

   Example2: induction:

              loop2:
              for (i=0; i<N; i++)
                 a[i] = i;  */

static void
vect_analyze_scalar_cycles (loop_vec_info loop_vinfo)
{
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo);

  vect_analyze_scalar_cycles_1 (loop_vinfo, loop);

  /* When vectorizing an outer-loop, the inner-loop is executed sequentially.
     Reductions in such inner-loop therefore have different properties than
     the reductions in the nest that gets vectorized:
     1. When vectorized, they are executed in the same order as in the original
        scalar loop, so we can't change the order of computation when
        vectorizing them.
     2. FIXME: Inner-loop reductions can be used in the inner-loop, so the
        current checks are too strict.  */

  if (loop->inner)
    vect_analyze_scalar_cycles_1 (loop_vinfo, loop->inner);
}

/* Transfer group and reduction information from STMT_INFO to its
   pattern stmt.  */

static void
vect_fixup_reduc_chain (stmt_vec_info stmt_info)
{
  stmt_vec_info firstp = STMT_VINFO_RELATED_STMT (stmt_info);
  stmt_vec_info stmtp;
  gcc_assert (!REDUC_GROUP_FIRST_ELEMENT (firstp)
	      && REDUC_GROUP_FIRST_ELEMENT (stmt_info));
  REDUC_GROUP_SIZE (firstp) = REDUC_GROUP_SIZE (stmt_info);
  do
    {
      stmtp = STMT_VINFO_RELATED_STMT (stmt_info);
      gcc_checking_assert (STMT_VINFO_DEF_TYPE (stmtp)
			   == STMT_VINFO_DEF_TYPE (stmt_info));
      REDUC_GROUP_FIRST_ELEMENT (stmtp) = firstp;
      stmt_info = REDUC_GROUP_NEXT_ELEMENT (stmt_info);
      if (stmt_info)
	REDUC_GROUP_NEXT_ELEMENT (stmtp)
	  = STMT_VINFO_RELATED_STMT (stmt_info);
    }
  while (stmt_info);
}

/* Fixup scalar cycles that now have their stmts detected as patterns.  */

static void
vect_fixup_scalar_cycles_with_patterns (loop_vec_info loop_vinfo)
{
  stmt_vec_info first;
  unsigned i;

  FOR_EACH_VEC_ELT (LOOP_VINFO_REDUCTION_CHAINS (loop_vinfo), i, first)
    if (STMT_VINFO_IN_PATTERN_P (first))
      {
	stmt_vec_info next = REDUC_GROUP_NEXT_ELEMENT (first);
	while (next)
	  {
	    if (! STMT_VINFO_IN_PATTERN_P (next)
		|| STMT_VINFO_REDUC_IDX (STMT_VINFO_RELATED_STMT (next)) == -1)
	      break;
	    next = REDUC_GROUP_NEXT_ELEMENT (next);
	  }
	/* If not all stmt in the chain are patterns or if we failed
	   to update STMT_VINFO_REDUC_IDX try to handle the chain
	   without patterns.  */
	if (! next
	    && STMT_VINFO_REDUC_IDX (STMT_VINFO_RELATED_STMT (first)) != -1)
	  {
	    vect_fixup_reduc_chain (first);
	    LOOP_VINFO_REDUCTION_CHAINS (loop_vinfo)[i]
	      = STMT_VINFO_RELATED_STMT (first);
	  }
      }
}

/* Function vect_get_loop_niters.

   Determine how many iterations the loop is executed and place it
   in NUMBER_OF_ITERATIONS.  Place the number of latch iterations
   in NUMBER_OF_ITERATIONSM1.  Place the condition under which the
   niter information holds in ASSUMPTIONS.

   Return the loop exit condition.  */


static gcond *
vect_get_loop_niters (class loop *loop, tree *assumptions,
		      tree *number_of_iterations, tree *number_of_iterationsm1)
{
  edge exit = single_exit (loop);
  class tree_niter_desc niter_desc;
  tree niter_assumptions, niter, may_be_zero;
  gcond *cond = get_loop_exit_condition (loop);

  *assumptions = boolean_true_node;
  *number_of_iterationsm1 = chrec_dont_know;
  *number_of_iterations = chrec_dont_know;
  DUMP_VECT_SCOPE ("get_loop_niters");

  if (!exit)
    return cond;

  may_be_zero = NULL_TREE;
  if (!number_of_iterations_exit_assumptions (loop, exit, &niter_desc, NULL)
      || chrec_contains_undetermined (niter_desc.niter))
    return cond;

  niter_assumptions = niter_desc.assumptions;
  may_be_zero = niter_desc.may_be_zero;
  niter = niter_desc.niter;

  if (may_be_zero && integer_zerop (may_be_zero))
    may_be_zero = NULL_TREE;

  if (may_be_zero)
    {
      if (COMPARISON_CLASS_P (may_be_zero))
	{
	  /* Try to combine may_be_zero with assumptions, this can simplify
	     computation of niter expression.  */
	  if (niter_assumptions && !integer_nonzerop (niter_assumptions))
	    niter_assumptions = fold_build2 (TRUTH_AND_EXPR, boolean_type_node,
					     niter_assumptions,
					     fold_build1 (TRUTH_NOT_EXPR,
							  boolean_type_node,
							  may_be_zero));
	  else
	    niter = fold_build3 (COND_EXPR, TREE_TYPE (niter), may_be_zero,
				 build_int_cst (TREE_TYPE (niter), 0),
				 rewrite_to_non_trapping_overflow (niter));

	  may_be_zero = NULL_TREE;
	}
      else if (integer_nonzerop (may_be_zero))
	{
	  *number_of_iterationsm1 = build_int_cst (TREE_TYPE (niter), 0);
	  *number_of_iterations = build_int_cst (TREE_TYPE (niter), 1);
	  return cond;
	}
      else
	return cond;
    }

  *assumptions = niter_assumptions;
  *number_of_iterationsm1 = niter;

  /* We want the number of loop header executions which is the number
     of latch executions plus one.
     ???  For UINT_MAX latch executions this number overflows to zero
     for loops like do { n++; } while (n != 0);  */
  if (niter && !chrec_contains_undetermined (niter))
    niter = fold_build2 (PLUS_EXPR, TREE_TYPE (niter), unshare_expr (niter),
			  build_int_cst (TREE_TYPE (niter), 1));
  *number_of_iterations = niter;

  return cond;
}

/* Function bb_in_loop_p

   Used as predicate for dfs order traversal of the loop bbs.  */

static bool
bb_in_loop_p (const_basic_block bb, const void *data)
{
  const class loop *const loop = (const class loop *)data;
  if (flow_bb_inside_loop_p (loop, bb))
    return true;
  return false;
}


/* Create and initialize a new loop_vec_info struct for LOOP_IN, as well as
   stmt_vec_info structs for all the stmts in LOOP_IN.  */

_loop_vec_info::_loop_vec_info (class loop *loop_in, vec_info_shared *shared)
  : vec_info (vec_info::loop, init_cost (loop_in), shared),
    loop (loop_in),
    bbs (XCNEWVEC (basic_block, loop->num_nodes)),
    num_itersm1 (NULL_TREE),
    num_iters (NULL_TREE),
    num_iters_unchanged (NULL_TREE),
    num_iters_assumptions (NULL_TREE),
    th (0),
    versioning_threshold (0),
    vectorization_factor (0),
    max_vectorization_factor (0),
    mask_skip_niters (NULL_TREE),
    mask_compare_type (NULL_TREE),
    simd_if_cond (NULL_TREE),
    unaligned_dr (NULL),
    peeling_for_alignment (0),
    ptr_mask (0),
    ivexpr_map (NULL),
    scan_map (NULL),
    slp_unrolling_factor (1),
    single_scalar_iteration_cost (0),
    vec_outside_cost (0),
    vec_inside_cost (0),
    vectorizable (false),
    can_fully_mask_p (true),
    fully_masked_p (false),
    peeling_for_gaps (false),
    peeling_for_niter (false),
    no_data_dependencies (false),
    has_mask_store (false),
    scalar_loop_scaling (profile_probability::uninitialized ()),
    scalar_loop (NULL),
    orig_loop_info (NULL)
{
  /* CHECKME: We want to visit all BBs before their successors (except for
     latch blocks, for which this assertion wouldn't hold).  In the simple
     case of the loop forms we allow, a dfs order of the BBs would the same
     as reversed postorder traversal, so we are safe.  */

  unsigned int nbbs = dfs_enumerate_from (loop->header, 0, bb_in_loop_p,
					  bbs, loop->num_nodes, loop);
  gcc_assert (nbbs == loop->num_nodes);

  for (unsigned int i = 0; i < nbbs; i++)
    {
      basic_block bb = bbs[i];
      gimple_stmt_iterator si;

      for (si = gsi_start_phis (bb); !gsi_end_p (si); gsi_next (&si))
	{
	  gimple *phi = gsi_stmt (si);
	  gimple_set_uid (phi, 0);
	  add_stmt (phi);
	}

      for (si = gsi_start_bb (bb); !gsi_end_p (si); gsi_next (&si))
	{
	  gimple *stmt = gsi_stmt (si);
	  gimple_set_uid (stmt, 0);
	  add_stmt (stmt);
	  /* If .GOMP_SIMD_LANE call for the current loop has 3 arguments, the
	     third argument is the #pragma omp simd if (x) condition, when 0,
	     loop shouldn't be vectorized, when non-zero constant, it should
	     be vectorized normally, otherwise versioned with vectorized loop
	     done if the condition is non-zero at runtime.  */
	  if (loop_in->simduid
	      && is_gimple_call (stmt)
	      && gimple_call_internal_p (stmt)
	      && gimple_call_internal_fn (stmt) == IFN_GOMP_SIMD_LANE
	      && gimple_call_num_args (stmt) >= 3
	      && TREE_CODE (gimple_call_arg (stmt, 0)) == SSA_NAME
	      && (loop_in->simduid
		  == SSA_NAME_VAR (gimple_call_arg (stmt, 0))))
	    {
	      tree arg = gimple_call_arg (stmt, 2);
	      if (integer_zerop (arg) || TREE_CODE (arg) == SSA_NAME)
		simd_if_cond = arg;
	      else
		gcc_assert (integer_nonzerop (arg));
	    }
	}
    }

  epilogue_vinfos.create (6);
}

/* Free all levels of MASKS.  */

void
release_vec_loop_masks (vec_loop_masks *masks)
{
  rgroup_masks *rgm;
  unsigned int i;
  FOR_EACH_VEC_ELT (*masks, i, rgm)
    rgm->masks.release ();
  masks->release ();
}

/* Free all memory used by the _loop_vec_info, as well as all the
   stmt_vec_info structs of all the stmts in the loop.  */

_loop_vec_info::~_loop_vec_info ()
{
  free (bbs);

  release_vec_loop_masks (&masks);
  delete ivexpr_map;
  delete scan_map;
  epilogue_vinfos.release ();

  loop->aux = NULL;
}

/* Return an invariant or register for EXPR and emit necessary
   computations in the LOOP_VINFO loop preheader.  */

tree
cse_and_gimplify_to_preheader (loop_vec_info loop_vinfo, tree expr)
{
  if (is_gimple_reg (expr)
      || is_gimple_min_invariant (expr))
    return expr;

  if (! loop_vinfo->ivexpr_map)
    loop_vinfo->ivexpr_map = new hash_map<tree_operand_hash, tree>;
  tree &cached = loop_vinfo->ivexpr_map->get_or_insert (expr);
  if (! cached)
    {
      gimple_seq stmts = NULL;
      cached = force_gimple_operand (unshare_expr (expr),
				     &stmts, true, NULL_TREE);
      if (stmts)
	{
	  edge e = loop_preheader_edge (LOOP_VINFO_LOOP (loop_vinfo));
	  gsi_insert_seq_on_edge_immediate (e, stmts);
	}
    }
  return cached;
}

/* Return true if we can use CMP_TYPE as the comparison type to produce
   all masks required to mask LOOP_VINFO.  */

static bool
can_produce_all_loop_masks_p (loop_vec_info loop_vinfo, tree cmp_type)
{
  rgroup_masks *rgm;
  unsigned int i;
  FOR_EACH_VEC_ELT (LOOP_VINFO_MASKS (loop_vinfo), i, rgm)
    if (rgm->mask_type != NULL_TREE
	&& !direct_internal_fn_supported_p (IFN_WHILE_ULT,
					    cmp_type, rgm->mask_type,
					    OPTIMIZE_FOR_SPEED))
      return false;
  return true;
}

/* Calculate the maximum number of scalars per iteration for every
   rgroup in LOOP_VINFO.  */

static unsigned int
vect_get_max_nscalars_per_iter (loop_vec_info loop_vinfo)
{
  unsigned int res = 1;
  unsigned int i;
  rgroup_masks *rgm;
  FOR_EACH_VEC_ELT (LOOP_VINFO_MASKS (loop_vinfo), i, rgm)
    res = MAX (res, rgm->max_nscalars_per_iter);
  return res;
}

/* Each statement in LOOP_VINFO can be masked where necessary.  Check
   whether we can actually generate the masks required.  Return true if so,
   storing the type of the scalar IV in LOOP_VINFO_MASK_COMPARE_TYPE.  */

static bool
vect_verify_full_masking (loop_vec_info loop_vinfo)
{
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo);
  unsigned int min_ni_width;
  unsigned int max_nscalars_per_iter
    = vect_get_max_nscalars_per_iter (loop_vinfo);

  /* Use a normal loop if there are no statements that need masking.
     This only happens in rare degenerate cases: it means that the loop
     has no loads, no stores, and no live-out values.  */
  if (LOOP_VINFO_MASKS (loop_vinfo).is_empty ())
    return false;

  /* Get the maximum number of iterations that is representable
     in the counter type.  */
  tree ni_type = TREE_TYPE (LOOP_VINFO_NITERSM1 (loop_vinfo));
  widest_int max_ni = wi::to_widest (TYPE_MAX_VALUE (ni_type)) + 1;

  /* Get a more refined estimate for the number of iterations.  */
  widest_int max_back_edges;
  if (max_loop_iterations (loop, &max_back_edges))
    max_ni = wi::smin (max_ni, max_back_edges + 1);

  /* Account for rgroup masks, in which each bit is replicated N times.  */
  max_ni *= max_nscalars_per_iter;

  /* Work out how many bits we need to represent the limit.  */
  min_ni_width = wi::min_precision (max_ni, UNSIGNED);

  /* Find a scalar mode for which WHILE_ULT is supported.  */
  opt_scalar_int_mode cmp_mode_iter;
  tree cmp_type = NULL_TREE;
  tree iv_type = NULL_TREE;
  widest_int iv_limit = vect_iv_limit_for_full_masking (loop_vinfo);
  unsigned int iv_precision = UINT_MAX;

  if (iv_limit != -1)
    iv_precision = wi::min_precision (iv_limit * max_nscalars_per_iter,
				      UNSIGNED);

  FOR_EACH_MODE_IN_CLASS (cmp_mode_iter, MODE_INT)
    {
      unsigned int cmp_bits = GET_MODE_BITSIZE (cmp_mode_iter.require ());
      if (cmp_bits >= min_ni_width
	  && targetm.scalar_mode_supported_p (cmp_mode_iter.require ()))
	{
	  tree this_type = build_nonstandard_integer_type (cmp_bits, true);
	  if (this_type
	      && can_produce_all_loop_masks_p (loop_vinfo, this_type))
	    {
	      /* Although we could stop as soon as we find a valid mode,
		 there are at least two reasons why that's not always the
		 best choice:

		 - An IV that's Pmode or wider is more likely to be reusable
		   in address calculations than an IV that's narrower than
		   Pmode.

		 - Doing the comparison in IV_PRECISION or wider allows
		   a natural 0-based IV, whereas using a narrower comparison
		   type requires mitigations against wrap-around.

		 Conversely, if the IV limit is variable, doing the comparison
		 in a wider type than the original type can introduce
		 unnecessary extensions, so picking the widest valid mode
		 is not always a good choice either.

		 Here we prefer the first IV type that's Pmode or wider,
		 and the first comparison type that's IV_PRECISION or wider.
		 (The comparison type must be no wider than the IV type,
		 to avoid extensions in the vector loop.)

		 ??? We might want to try continuing beyond Pmode for ILP32
		 targets if CMP_BITS < IV_PRECISION.  */
	      iv_type = this_type;
	      if (!cmp_type || iv_precision > TYPE_PRECISION (cmp_type))
		cmp_type = this_type;
	      if (cmp_bits >= GET_MODE_BITSIZE (Pmode))
		break;
	    }
	}
    }

  if (!cmp_type)
    return false;

  LOOP_VINFO_MASK_COMPARE_TYPE (loop_vinfo) = cmp_type;
  LOOP_VINFO_MASK_IV_TYPE (loop_vinfo) = iv_type;
  return true;
}

/* Calculate the cost of one scalar iteration of the loop.  */
static void
vect_compute_single_scalar_iteration_cost (loop_vec_info loop_vinfo)
{
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo);
  basic_block *bbs = LOOP_VINFO_BBS (loop_vinfo);
  int nbbs = loop->num_nodes, factor;
  int innerloop_iters, i;

  DUMP_VECT_SCOPE ("vect_compute_single_scalar_iteration_cost");

  /* Gather costs for statements in the scalar loop.  */

  /* FORNOW.  */
  innerloop_iters = 1;
  if (loop->inner)
    innerloop_iters = 50; /* FIXME */

  for (i = 0; i < nbbs; i++)
    {
      gimple_stmt_iterator si;
      basic_block bb = bbs[i];

      if (bb->loop_father == loop->inner)
        factor = innerloop_iters;
      else
        factor = 1;

      for (si = gsi_start_bb (bb); !gsi_end_p (si); gsi_next (&si))
        {
	  gimple *stmt = gsi_stmt (si);
	  stmt_vec_info stmt_info = loop_vinfo->lookup_stmt (stmt);

          if (!is_gimple_assign (stmt) && !is_gimple_call (stmt))
            continue;

          /* Skip stmts that are not vectorized inside the loop.  */
	  stmt_vec_info vstmt_info = vect_stmt_to_vectorize (stmt_info);
          if (!STMT_VINFO_RELEVANT_P (vstmt_info)
              && (!STMT_VINFO_LIVE_P (vstmt_info)
                  || !VECTORIZABLE_CYCLE_DEF
			(STMT_VINFO_DEF_TYPE (vstmt_info))))
            continue;

	  vect_cost_for_stmt kind;
          if (STMT_VINFO_DATA_REF (stmt_info))
            {
              if (DR_IS_READ (STMT_VINFO_DATA_REF (stmt_info)))
               kind = scalar_load;
             else
               kind = scalar_store;
            }
	  else if (vect_nop_conversion_p (stmt_info))
	    continue;
	  else
            kind = scalar_stmt;

	  record_stmt_cost (&LOOP_VINFO_SCALAR_ITERATION_COST (loop_vinfo),
			    factor, kind, stmt_info, 0, vect_prologue);
        }
    }

  /* Now accumulate cost.  */
  void *target_cost_data = init_cost (loop);
  stmt_info_for_cost *si;
  int j;
  FOR_EACH_VEC_ELT (LOOP_VINFO_SCALAR_ITERATION_COST (loop_vinfo),
		    j, si)
    (void) add_stmt_cost (loop_vinfo, target_cost_data, si->count,
			  si->kind, si->stmt_info, si->vectype,
			  si->misalign, vect_body);
  unsigned dummy, body_cost = 0;
  finish_cost (target_cost_data, &dummy, &body_cost, &dummy);
  destroy_cost_data (target_cost_data);
  LOOP_VINFO_SINGLE_SCALAR_ITERATION_COST (loop_vinfo) = body_cost;
}


/* Function vect_analyze_loop_form_1.

   Verify that certain CFG restrictions hold, including:
   - the loop has a pre-header
   - the loop has a single entry and exit
   - the loop exit condition is simple enough
   - the number of iterations can be analyzed, i.e, a countable loop.  The
     niter could be analyzed under some assumptions.  */

opt_result
vect_analyze_loop_form_1 (class loop *loop, gcond **loop_cond,
			  tree *assumptions, tree *number_of_iterationsm1,
			  tree *number_of_iterations, gcond **inner_loop_cond)
{
  DUMP_VECT_SCOPE ("vect_analyze_loop_form");

  /* Different restrictions apply when we are considering an inner-most loop,
     vs. an outer (nested) loop.
     (FORNOW. May want to relax some of these restrictions in the future).  */

  if (!loop->inner)
    {
      /* Inner-most loop.  We currently require that the number of BBs is
	 exactly 2 (the header and latch).  Vectorizable inner-most loops
	 look like this:

                        (pre-header)
                           |
                          header <--------+
                           | |            |
                           | +--> latch --+
                           |
                        (exit-bb)  */

      if (loop->num_nodes != 2)
	return opt_result::failure_at (vect_location,
				       "not vectorized:"
				       " control flow in loop.\n");

      if (empty_block_p (loop->header))
	return opt_result::failure_at (vect_location,
				       "not vectorized: empty loop.\n");
    }
  else
    {
      class loop *innerloop = loop->inner;
      edge entryedge;

      /* Nested loop. We currently require that the loop is doubly-nested,
	 contains a single inner loop, and the number of BBs is exactly 5.
	 Vectorizable outer-loops look like this:

			(pre-header)
			   |
			  header <---+
			   |         |
		          inner-loop |
			   |         |
			  tail ------+
			   |
		        (exit-bb)

	 The inner-loop has the properties expected of inner-most loops
	 as described above.  */

      if ((loop->inner)->inner || (loop->inner)->next)
	return opt_result::failure_at (vect_location,
				       "not vectorized:"
				       " multiple nested loops.\n");

      if (loop->num_nodes != 5)
	return opt_result::failure_at (vect_location,
				       "not vectorized:"
				       " control flow in loop.\n");

      entryedge = loop_preheader_edge (innerloop);
      if (entryedge->src != loop->header
	  || !single_exit (innerloop)
	  || single_exit (innerloop)->dest != EDGE_PRED (loop->latch, 0)->src)
	return opt_result::failure_at (vect_location,
				       "not vectorized:"
				       " unsupported outerloop form.\n");

      /* Analyze the inner-loop.  */
      tree inner_niterm1, inner_niter, inner_assumptions;
      opt_result res
	= vect_analyze_loop_form_1 (loop->inner, inner_loop_cond,
				    &inner_assumptions, &inner_niterm1,
				    &inner_niter, NULL);
      if (!res)
	{
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			     "not vectorized: Bad inner loop.\n");
	  return res;
	}

      /* Don't support analyzing niter under assumptions for inner
	 loop.  */
      if (!integer_onep (inner_assumptions))
	return opt_result::failure_at (vect_location,
				       "not vectorized: Bad inner loop.\n");

      if (!expr_invariant_in_loop_p (loop, inner_niter))
	return opt_result::failure_at (vect_location,
				       "not vectorized: inner-loop count not"
				       " invariant.\n");

      if (dump_enabled_p ())
        dump_printf_loc (MSG_NOTE, vect_location,
			 "Considering outer-loop vectorization.\n");
    }

  if (!single_exit (loop))
    return opt_result::failure_at (vect_location,
				   "not vectorized: multiple exits.\n");
  if (EDGE_COUNT (loop->header->preds) != 2)
    return opt_result::failure_at (vect_location,
				   "not vectorized:"
				   " too many incoming edges.\n");

  /* We assume that the loop exit condition is at the end of the loop. i.e,
     that the loop is represented as a do-while (with a proper if-guard
     before the loop if needed), where the loop header contains all the
     executable statements, and the latch is empty.  */
  if (!empty_block_p (loop->latch)
      || !gimple_seq_empty_p (phi_nodes (loop->latch)))
    return opt_result::failure_at (vect_location,
				   "not vectorized: latch block not empty.\n");

  /* Make sure the exit is not abnormal.  */
  edge e = single_exit (loop);
  if (e->flags & EDGE_ABNORMAL)
    return opt_result::failure_at (vect_location,
				   "not vectorized:"
				   " abnormal loop exit edge.\n");

  *loop_cond = vect_get_loop_niters (loop, assumptions, number_of_iterations,
				     number_of_iterationsm1);
  if (!*loop_cond)
    return opt_result::failure_at
      (vect_location,
       "not vectorized: complicated exit condition.\n");

  if (integer_zerop (*assumptions)
      || !*number_of_iterations
      || chrec_contains_undetermined (*number_of_iterations))
    return opt_result::failure_at
      (*loop_cond,
       "not vectorized: number of iterations cannot be computed.\n");

  if (integer_zerop (*number_of_iterations))
    return opt_result::failure_at
      (*loop_cond,
       "not vectorized: number of iterations = 0.\n");

  return opt_result::success ();
}

/* Analyze LOOP form and return a loop_vec_info if it is of suitable form.  */

opt_loop_vec_info
vect_analyze_loop_form (class loop *loop, vec_info_shared *shared)
{
  tree assumptions, number_of_iterations, number_of_iterationsm1;
  gcond *loop_cond, *inner_loop_cond = NULL;

  opt_result res
    = vect_analyze_loop_form_1 (loop, &loop_cond,
				&assumptions, &number_of_iterationsm1,
				&number_of_iterations, &inner_loop_cond);
  if (!res)
    return opt_loop_vec_info::propagate_failure (res);

  loop_vec_info loop_vinfo = new _loop_vec_info (loop, shared);
  LOOP_VINFO_NITERSM1 (loop_vinfo) = number_of_iterationsm1;
  LOOP_VINFO_NITERS (loop_vinfo) = number_of_iterations;
  LOOP_VINFO_NITERS_UNCHANGED (loop_vinfo) = number_of_iterations;
  if (!integer_onep (assumptions))
    {
      /* We consider to vectorize this loop by versioning it under
	 some assumptions.  In order to do this, we need to clear
	 existing information computed by scev and niter analyzer.  */
      scev_reset_htab ();
      free_numbers_of_iterations_estimates (loop);
      /* Also set flag for this loop so that following scev and niter
	 analysis are done under the assumptions.  */
      loop_constraint_set (loop, LOOP_C_FINITE);
      /* Also record the assumptions for versioning.  */
      LOOP_VINFO_NITERS_ASSUMPTIONS (loop_vinfo) = assumptions;
    }

  if (!LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo))
    {
      if (dump_enabled_p ())
        {
          dump_printf_loc (MSG_NOTE, vect_location,
			   "Symbolic number of iterations is ");
	  dump_generic_expr (MSG_NOTE, TDF_DETAILS, number_of_iterations);
          dump_printf (MSG_NOTE, "\n");
        }
    }

  stmt_vec_info loop_cond_info = loop_vinfo->lookup_stmt (loop_cond);
  STMT_VINFO_TYPE (loop_cond_info) = loop_exit_ctrl_vec_info_type;
  if (inner_loop_cond)
    {
      stmt_vec_info inner_loop_cond_info
	= loop_vinfo->lookup_stmt (inner_loop_cond);
      STMT_VINFO_TYPE (inner_loop_cond_info) = loop_exit_ctrl_vec_info_type;
    }

  gcc_assert (!loop->aux);
  loop->aux = loop_vinfo;
  return opt_loop_vec_info::success (loop_vinfo);
}



/* Scan the loop stmts and dependent on whether there are any (non-)SLP
   statements update the vectorization factor.  */

static void
vect_update_vf_for_slp (loop_vec_info loop_vinfo)
{
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo);
  basic_block *bbs = LOOP_VINFO_BBS (loop_vinfo);
  int nbbs = loop->num_nodes;
  poly_uint64 vectorization_factor;
  int i;

  DUMP_VECT_SCOPE ("vect_update_vf_for_slp");

  vectorization_factor = LOOP_VINFO_VECT_FACTOR (loop_vinfo);
  gcc_assert (known_ne (vectorization_factor, 0U));

  /* If all the stmts in the loop can be SLPed, we perform only SLP, and
     vectorization factor of the loop is the unrolling factor required by
     the SLP instances.  If that unrolling factor is 1, we say, that we
     perform pure SLP on loop - cross iteration parallelism is not
     exploited.  */
  bool only_slp_in_loop = true;
  for (i = 0; i < nbbs; i++)
    {
      basic_block bb = bbs[i];
      for (gphi_iterator si = gsi_start_phis (bb); !gsi_end_p (si);
	   gsi_next (&si))
	{
	  stmt_vec_info stmt_info = loop_vinfo->lookup_stmt (si.phi ());
	  if (!stmt_info)
	    continue;
	  if ((STMT_VINFO_RELEVANT_P (stmt_info)
	       || VECTORIZABLE_CYCLE_DEF (STMT_VINFO_DEF_TYPE (stmt_info)))
	      && !PURE_SLP_STMT (stmt_info))
	    /* STMT needs both SLP and loop-based vectorization.  */
	    only_slp_in_loop = false;
	}
      for (gimple_stmt_iterator si = gsi_start_bb (bb); !gsi_end_p (si);
	   gsi_next (&si))
	{
	  stmt_vec_info stmt_info = loop_vinfo->lookup_stmt (gsi_stmt (si));
	  stmt_info = vect_stmt_to_vectorize (stmt_info);
	  if ((STMT_VINFO_RELEVANT_P (stmt_info)
	       || VECTORIZABLE_CYCLE_DEF (STMT_VINFO_DEF_TYPE (stmt_info)))
	      && !PURE_SLP_STMT (stmt_info))
	    /* STMT needs both SLP and loop-based vectorization.  */
	    only_slp_in_loop = false;
	}
    }

  if (only_slp_in_loop)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location,
			 "Loop contains only SLP stmts\n");
      vectorization_factor = LOOP_VINFO_SLP_UNROLLING_FACTOR (loop_vinfo);
    }
  else
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location,
			 "Loop contains SLP and non-SLP stmts\n");
      /* Both the vectorization factor and unroll factor have the form
	 GET_MODE_SIZE (loop_vinfo->vector_mode) * X for some rational X,
	 so they must have a common multiple.  */
      vectorization_factor
	= force_common_multiple (vectorization_factor,
				 LOOP_VINFO_SLP_UNROLLING_FACTOR (loop_vinfo));
    }

  LOOP_VINFO_VECT_FACTOR (loop_vinfo) = vectorization_factor;
  if (dump_enabled_p ())
    {
      dump_printf_loc (MSG_NOTE, vect_location,
		       "Updating vectorization factor to ");
      dump_dec (MSG_NOTE, vectorization_factor);
      dump_printf (MSG_NOTE, ".\n");
    }
}

/* Return true if STMT_INFO describes a double reduction phi and if
   the other phi in the reduction is also relevant for vectorization.
   This rejects cases such as:

      outer1:
	x_1 = PHI <x_3(outer2), ...>;
	...

      inner:
	x_2 = ...;
	...

      outer2:
	x_3 = PHI <x_2(inner)>;

   if nothing in x_2 or elsewhere makes x_1 relevant.  */

static bool
vect_active_double_reduction_p (stmt_vec_info stmt_info)
{
  if (STMT_VINFO_DEF_TYPE (stmt_info) != vect_double_reduction_def)
    return false;

  return STMT_VINFO_RELEVANT_P (STMT_VINFO_REDUC_DEF (stmt_info));
}

/* Function vect_analyze_loop_operations.

   Scan the loop stmts and make sure they are all vectorizable.  */

static opt_result
vect_analyze_loop_operations (loop_vec_info loop_vinfo)
{
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo);
  basic_block *bbs = LOOP_VINFO_BBS (loop_vinfo);
  int nbbs = loop->num_nodes;
  int i;
  stmt_vec_info stmt_info;
  bool need_to_vectorize = false;
  bool ok;

  DUMP_VECT_SCOPE ("vect_analyze_loop_operations");

  auto_vec<stmt_info_for_cost> cost_vec;

  for (i = 0; i < nbbs; i++)
    {
      basic_block bb = bbs[i];

      for (gphi_iterator si = gsi_start_phis (bb); !gsi_end_p (si);
	   gsi_next (&si))
        {
          gphi *phi = si.phi ();
          ok = true;

	  stmt_info = loop_vinfo->lookup_stmt (phi);
          if (dump_enabled_p ())
	    dump_printf_loc (MSG_NOTE, vect_location, "examining phi: %G", phi);
	  if (virtual_operand_p (gimple_phi_result (phi)))
	    continue;

          /* Inner-loop loop-closed exit phi in outer-loop vectorization
             (i.e., a phi in the tail of the outer-loop).  */
          if (! is_loop_header_bb_p (bb))
            {
              /* FORNOW: we currently don't support the case that these phis
                 are not used in the outerloop (unless it is double reduction,
                 i.e., this phi is vect_reduction_def), cause this case
                 requires to actually do something here.  */
              if (STMT_VINFO_LIVE_P (stmt_info)
		  && !vect_active_double_reduction_p (stmt_info))
		return opt_result::failure_at (phi,
					       "Unsupported loop-closed phi"
					       " in outer-loop.\n");

              /* If PHI is used in the outer loop, we check that its operand
                 is defined in the inner loop.  */
              if (STMT_VINFO_RELEVANT_P (stmt_info))
                {
                  tree phi_op;

                  if (gimple_phi_num_args (phi) != 1)
                    return opt_result::failure_at (phi, "unsupported phi");

                  phi_op = PHI_ARG_DEF (phi, 0);
		  stmt_vec_info op_def_info = loop_vinfo->lookup_def (phi_op);
		  if (!op_def_info)
		    return opt_result::failure_at (phi, "unsupported phi\n");

		  if (STMT_VINFO_RELEVANT (op_def_info) != vect_used_in_outer
		      && (STMT_VINFO_RELEVANT (op_def_info)
			  != vect_used_in_outer_by_reduction))
		    return opt_result::failure_at (phi, "unsupported phi\n");

		  if ((STMT_VINFO_DEF_TYPE (stmt_info) == vect_internal_def
		       || (STMT_VINFO_DEF_TYPE (stmt_info)
			   == vect_double_reduction_def))
		      && !vectorizable_lc_phi (loop_vinfo,
					       stmt_info, NULL, NULL))
		    return opt_result::failure_at (phi, "unsupported phi\n");
                }

              continue;
            }

          gcc_assert (stmt_info);

          if ((STMT_VINFO_RELEVANT (stmt_info) == vect_used_in_scope
               || STMT_VINFO_LIVE_P (stmt_info))
              && STMT_VINFO_DEF_TYPE (stmt_info) != vect_induction_def)
	    /* A scalar-dependence cycle that we don't support.  */
	    return opt_result::failure_at (phi,
					   "not vectorized:"
					   " scalar dependence cycle.\n");

          if (STMT_VINFO_RELEVANT_P (stmt_info))
            {
              need_to_vectorize = true;
              if (STMT_VINFO_DEF_TYPE (stmt_info) == vect_induction_def
		  && ! PURE_SLP_STMT (stmt_info))
		ok = vectorizable_induction (loop_vinfo,
					     stmt_info, NULL, NULL, NULL,
					     &cost_vec);
	      else if ((STMT_VINFO_DEF_TYPE (stmt_info) == vect_reduction_def
			|| (STMT_VINFO_DEF_TYPE (stmt_info)
			    == vect_double_reduction_def)
			|| STMT_VINFO_DEF_TYPE (stmt_info) == vect_nested_cycle)
		       && ! PURE_SLP_STMT (stmt_info))
		ok = vectorizable_reduction (loop_vinfo,
					     stmt_info, NULL, NULL, &cost_vec);
            }

	  /* SLP PHIs are tested by vect_slp_analyze_node_operations.  */
	  if (ok
	      && STMT_VINFO_LIVE_P (stmt_info)
	      && !PURE_SLP_STMT (stmt_info))
	    ok = vectorizable_live_operation (loop_vinfo,
					      stmt_info, NULL, NULL, NULL,
					      -1, false, &cost_vec);

          if (!ok)
	    return opt_result::failure_at (phi,
					   "not vectorized: relevant phi not "
					   "supported: %G",
					   static_cast <gimple *> (phi));
        }

      for (gimple_stmt_iterator si = gsi_start_bb (bb); !gsi_end_p (si);
	   gsi_next (&si))
        {
	  gimple *stmt = gsi_stmt (si);
	  if (!gimple_clobber_p (stmt))
	    {
	      opt_result res
		= vect_analyze_stmt (loop_vinfo,
				     loop_vinfo->lookup_stmt (stmt),
				     &need_to_vectorize,
				     NULL, NULL, &cost_vec);
	      if (!res)
		return res;
	    }
        }
    } /* bbs */

  add_stmt_costs (loop_vinfo, loop_vinfo->target_cost_data, &cost_vec);

  /* All operations in the loop are either irrelevant (deal with loop
     control, or dead), or only used outside the loop and can be moved
     out of the loop (e.g. invariants, inductions).  The loop can be
     optimized away by scalar optimizations.  We're better off not
     touching this loop.  */
  if (!need_to_vectorize)
    {
      if (dump_enabled_p ())
        dump_printf_loc (MSG_NOTE, vect_location,
			 "All the computation can be taken out of the loop.\n");
      return opt_result::failure_at
	(vect_location,
	 "not vectorized: redundant loop. no profit to vectorize.\n");
    }

  return opt_result::success ();
}

/* Analyze the cost of the loop described by LOOP_VINFO.  Decide if it
   is worthwhile to vectorize.  Return 1 if definitely yes, 0 if
   definitely no, or -1 if it's worth retrying.  */

static int
vect_analyze_loop_costing (loop_vec_info loop_vinfo)
{
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo);
  unsigned int assumed_vf = vect_vf_for_cost (loop_vinfo);

  /* Only fully-masked loops can have iteration counts less than the
     vectorization factor.  */
  if (!LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
    {
      HOST_WIDE_INT max_niter;

      if (LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo))
	max_niter = LOOP_VINFO_INT_NITERS (loop_vinfo);
      else
	max_niter = max_stmt_executions_int (loop);

      if (max_niter != -1
	  && (unsigned HOST_WIDE_INT) max_niter < assumed_vf)
	{
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			     "not vectorized: iteration count smaller than "
			     "vectorization factor.\n");
	  return 0;
	}
    }

  int min_profitable_iters, min_profitable_estimate;
  vect_estimate_min_profitable_iters (loop_vinfo, &min_profitable_iters,
				      &min_profitable_estimate);

  if (min_profitable_iters < 0)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "not vectorized: vectorization not profitable.\n");
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "not vectorized: vector version will never be "
			 "profitable.\n");
      return -1;
    }

  int min_scalar_loop_bound = (param_min_vect_loop_bound
			       * assumed_vf);

  /* Use the cost model only if it is more conservative than user specified
     threshold.  */
  unsigned int th = (unsigned) MAX (min_scalar_loop_bound,
				    min_profitable_iters);

  LOOP_VINFO_COST_MODEL_THRESHOLD (loop_vinfo) = th;

  if (LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo)
      && LOOP_VINFO_INT_NITERS (loop_vinfo) < th)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "not vectorized: vectorization not profitable.\n");
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location,
			 "not vectorized: iteration count smaller than user "
			 "specified loop bound parameter or minimum profitable "
			 "iterations (whichever is more conservative).\n");
      return 0;
    }

  /* The static profitablity threshold min_profitable_estimate includes
     the cost of having to check at runtime whether the scalar loop
     should be used instead.  If it turns out that we don't need or want
     such a check, the threshold we should use for the static estimate
     is simply the point at which the vector loop becomes more profitable
     than the scalar loop.  */
  if (min_profitable_estimate > min_profitable_iters
      && !LOOP_REQUIRES_VERSIONING (loop_vinfo)
      && !LOOP_VINFO_PEELING_FOR_NITER (loop_vinfo)
      && !LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo)
      && !vect_apply_runtime_profitability_check_p (loop_vinfo))
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location, "no need for a runtime"
			 " choice between the scalar and vector loops\n");
      min_profitable_estimate = min_profitable_iters;
    }

  HOST_WIDE_INT estimated_niter;

  /* If we are vectorizing an epilogue then we know the maximum number of
     scalar iterations it will cover is at least one lower than the
     vectorization factor of the main loop.  */
  if (LOOP_VINFO_EPILOGUE_P (loop_vinfo))
    estimated_niter
      = vect_vf_for_cost (LOOP_VINFO_ORIG_LOOP_INFO (loop_vinfo)) - 1;
  else
    {
      estimated_niter = estimated_stmt_executions_int (loop);
      if (estimated_niter == -1)
	estimated_niter = likely_max_stmt_executions_int (loop);
    }
  if (estimated_niter != -1
      && ((unsigned HOST_WIDE_INT) estimated_niter
	  < MAX (th, (unsigned) min_profitable_estimate)))
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "not vectorized: estimated iteration count too "
			 "small.\n");
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location,
			 "not vectorized: estimated iteration count smaller "
			 "than specified loop bound parameter or minimum "
			 "profitable iterations (whichever is more "
			 "conservative).\n");
      return -1;
    }

  return 1;
}

static opt_result
vect_get_datarefs_in_loop (loop_p loop, basic_block *bbs,
			   vec<data_reference_p> *datarefs,
			   unsigned int *n_stmts)
{
  *n_stmts = 0;
  for (unsigned i = 0; i < loop->num_nodes; i++)
    for (gimple_stmt_iterator gsi = gsi_start_bb (bbs[i]);
	 !gsi_end_p (gsi); gsi_next (&gsi))
      {
	gimple *stmt = gsi_stmt (gsi);
	if (is_gimple_debug (stmt))
	  continue;
	++(*n_stmts);
	opt_result res = vect_find_stmt_data_reference (loop, stmt, datarefs);
	if (!res)
	  {
	    if (is_gimple_call (stmt) && loop->safelen)
	      {
		tree fndecl = gimple_call_fndecl (stmt), op;
		if (fndecl != NULL_TREE)
		  {
		    cgraph_node *node = cgraph_node::get (fndecl);
		    if (node != NULL && node->simd_clones != NULL)
		      {
			unsigned int j, n = gimple_call_num_args (stmt);
			for (j = 0; j < n; j++)
			  {
			    op = gimple_call_arg (stmt, j);
			    if (DECL_P (op)
				|| (REFERENCE_CLASS_P (op)
				    && get_base_address (op)))
			      break;
			  }
			op = gimple_call_lhs (stmt);
			/* Ignore #pragma omp declare simd functions
			   if they don't have data references in the
			   call stmt itself.  */
			if (j == n
			    && !(op
				 && (DECL_P (op)
				     || (REFERENCE_CLASS_P (op)
					 && get_base_address (op)))))
			  continue;
		      }
		  }
	      }
	    return res;
	  }
	/* If dependence analysis will give up due to the limit on the
	   number of datarefs stop here and fail fatally.  */
	if (datarefs->length ()
	    > (unsigned)param_loop_max_datarefs_for_datadeps)
	  return opt_result::failure_at (stmt, "exceeded param "
					 "loop-max-datarefs-for-datadeps\n");
      }
  return opt_result::success ();
}

/* Look for SLP-only access groups and turn each individual access into its own
   group.  */
static void
vect_dissolve_slp_only_groups (loop_vec_info loop_vinfo)
{
  unsigned int i;
  struct data_reference *dr;

  DUMP_VECT_SCOPE ("vect_dissolve_slp_only_groups");

  vec<data_reference_p> datarefs = loop_vinfo->shared->datarefs;
  FOR_EACH_VEC_ELT (datarefs, i, dr)
    {
      gcc_assert (DR_REF (dr));
      stmt_vec_info stmt_info = loop_vinfo->lookup_stmt (DR_STMT (dr));

      /* Check if the load is a part of an interleaving chain.  */
      if (STMT_VINFO_GROUPED_ACCESS (stmt_info))
	{
	  stmt_vec_info first_element = DR_GROUP_FIRST_ELEMENT (stmt_info);
	  unsigned int group_size = DR_GROUP_SIZE (first_element);

	  /* Check if SLP-only groups.  */
	  if (!STMT_SLP_TYPE (stmt_info)
	      && STMT_VINFO_SLP_VECT_ONLY (first_element))
	    {
	      /* Dissolve the group.  */
	      STMT_VINFO_SLP_VECT_ONLY (first_element) = false;

	      stmt_vec_info vinfo = first_element;
	      while (vinfo)
		{
		  stmt_vec_info next = DR_GROUP_NEXT_ELEMENT (vinfo);
		  DR_GROUP_FIRST_ELEMENT (vinfo) = vinfo;
		  DR_GROUP_NEXT_ELEMENT (vinfo) = NULL;
		  DR_GROUP_SIZE (vinfo) = 1;
		  if (STMT_VINFO_STRIDED_P (first_element))
		    DR_GROUP_GAP (vinfo) = 0;
		  else
		    DR_GROUP_GAP (vinfo) = group_size - 1;
		  vinfo = next;
		}
	    }
	}
    }
}


/* Decides whether we need to create an epilogue loop to handle
   remaining scalar iterations and sets PEELING_FOR_NITERS accordingly.  */

void
determine_peel_for_niter (loop_vec_info loop_vinfo)
{
  LOOP_VINFO_PEELING_FOR_NITER (loop_vinfo) = false;

  unsigned HOST_WIDE_INT const_vf;
  HOST_WIDE_INT max_niter
    = likely_max_stmt_executions_int (LOOP_VINFO_LOOP (loop_vinfo));

  unsigned th = LOOP_VINFO_COST_MODEL_THRESHOLD (loop_vinfo);
  if (!th && LOOP_VINFO_ORIG_LOOP_INFO (loop_vinfo))
    th = LOOP_VINFO_COST_MODEL_THRESHOLD (LOOP_VINFO_ORIG_LOOP_INFO
					  (loop_vinfo));

  if (LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
    /* The main loop handles all iterations.  */
    LOOP_VINFO_PEELING_FOR_NITER (loop_vinfo) = false;
  else if (LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo)
	   && LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo) >= 0)
    {
      /* Work out the (constant) number of iterations that need to be
	 peeled for reasons other than niters.  */
      unsigned int peel_niter = LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo);
      if (LOOP_VINFO_PEELING_FOR_GAPS (loop_vinfo))
	peel_niter += 1;
      if (!multiple_p (LOOP_VINFO_INT_NITERS (loop_vinfo) - peel_niter,
		       LOOP_VINFO_VECT_FACTOR (loop_vinfo)))
	LOOP_VINFO_PEELING_FOR_NITER (loop_vinfo) = true;
    }
  else if (LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo)
	   /* ??? When peeling for gaps but not alignment, we could
	      try to check whether the (variable) niters is known to be
	      VF * N + 1.  That's something of a niche case though.  */
	   || LOOP_VINFO_PEELING_FOR_GAPS (loop_vinfo)
	   || !LOOP_VINFO_VECT_FACTOR (loop_vinfo).is_constant (&const_vf)
	   || ((tree_ctz (LOOP_VINFO_NITERS (loop_vinfo))
		< (unsigned) exact_log2 (const_vf))
	       /* In case of versioning, check if the maximum number of
		  iterations is greater than th.  If they are identical,
		  the epilogue is unnecessary.  */
	       && (!LOOP_REQUIRES_VERSIONING (loop_vinfo)
		   || ((unsigned HOST_WIDE_INT) max_niter
		       > (th / const_vf) * const_vf))))
    LOOP_VINFO_PEELING_FOR_NITER (loop_vinfo) = true;
}


/* Function vect_analyze_loop_2.

   Apply a set of analyses on LOOP, and create a loop_vec_info struct
   for it.  The different analyses will record information in the
   loop_vec_info struct.  */
static opt_result
vect_analyze_loop_2 (loop_vec_info loop_vinfo, bool &fatal, unsigned *n_stmts)
{
  opt_result ok = opt_result::success ();
  int res;
  unsigned int max_vf = MAX_VECTORIZATION_FACTOR;
  poly_uint64 min_vf = 2;
  loop_vec_info orig_loop_vinfo = NULL;

  /* If we are dealing with an epilogue then orig_loop_vinfo points to the
     loop_vec_info of the first vectorized loop.  */
  if (LOOP_VINFO_EPILOGUE_P (loop_vinfo))
    orig_loop_vinfo = LOOP_VINFO_ORIG_LOOP_INFO (loop_vinfo);
  else
    orig_loop_vinfo = loop_vinfo;
  gcc_assert (orig_loop_vinfo);

  /* The first group of checks is independent of the vector size.  */
  fatal = true;

  if (LOOP_VINFO_SIMD_IF_COND (loop_vinfo)
      && integer_zerop (LOOP_VINFO_SIMD_IF_COND (loop_vinfo)))
    return opt_result::failure_at (vect_location,
				   "not vectorized: simd if(0)\n");

  /* Find all data references in the loop (which correspond to vdefs/vuses)
     and analyze their evolution in the loop.  */

  loop_p loop = LOOP_VINFO_LOOP (loop_vinfo);

  /* Gather the data references and count stmts in the loop.  */
  if (!LOOP_VINFO_DATAREFS (loop_vinfo).exists ())
    {
      opt_result res
	= vect_get_datarefs_in_loop (loop, LOOP_VINFO_BBS (loop_vinfo),
				     &LOOP_VINFO_DATAREFS (loop_vinfo),
				     n_stmts);
      if (!res)
	{
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			     "not vectorized: loop contains function "
			     "calls or data references that cannot "
			     "be analyzed\n");
	  return res;
	}
      loop_vinfo->shared->save_datarefs ();
    }
  else
    loop_vinfo->shared->check_datarefs ();

  /* Analyze the data references and also adjust the minimal
     vectorization factor according to the loads and stores.  */

  ok = vect_analyze_data_refs (loop_vinfo, &min_vf, &fatal);
  if (!ok)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "bad data references.\n");
      return ok;
    }

  /* Classify all cross-iteration scalar data-flow cycles.
     Cross-iteration cycles caused by virtual phis are analyzed separately.  */
  vect_analyze_scalar_cycles (loop_vinfo);

  vect_pattern_recog (loop_vinfo);

  vect_fixup_scalar_cycles_with_patterns (loop_vinfo);

  /* Analyze the access patterns of the data-refs in the loop (consecutive,
     complex, etc.). FORNOW: Only handle consecutive access pattern.  */

  ok = vect_analyze_data_ref_accesses (loop_vinfo);
  if (!ok)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "bad data access.\n");
      return ok;
    }

  /* Data-flow analysis to detect stmts that do not need to be vectorized.  */

  ok = vect_mark_stmts_to_be_vectorized (loop_vinfo, &fatal);
  if (!ok)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "unexpected pattern.\n");
      return ok;
    }

  /* While the rest of the analysis below depends on it in some way.  */
  fatal = false;

  /* Analyze data dependences between the data-refs in the loop
     and adjust the maximum vectorization factor according to
     the dependences.
     FORNOW: fail at the first data dependence that we encounter.  */

  ok = vect_analyze_data_ref_dependences (loop_vinfo, &max_vf);
  if (!ok)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "bad data dependence.\n");
      return ok;
    }
  if (max_vf != MAX_VECTORIZATION_FACTOR
      && maybe_lt (max_vf, min_vf))
    return opt_result::failure_at (vect_location, "bad data dependence.\n");
  LOOP_VINFO_MAX_VECT_FACTOR (loop_vinfo) = max_vf;

  ok = vect_determine_vectorization_factor (loop_vinfo);
  if (!ok)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "can't determine vectorization factor.\n");
      return ok;
    }
  if (max_vf != MAX_VECTORIZATION_FACTOR
      && maybe_lt (max_vf, LOOP_VINFO_VECT_FACTOR (loop_vinfo)))
    return opt_result::failure_at (vect_location, "bad data dependence.\n");

  /* Compute the scalar iteration cost.  */
  vect_compute_single_scalar_iteration_cost (loop_vinfo);

  poly_uint64 saved_vectorization_factor = LOOP_VINFO_VECT_FACTOR (loop_vinfo);

  /* Check the SLP opportunities in the loop, analyze and build SLP trees.  */
  ok = vect_analyze_slp (loop_vinfo, *n_stmts);
  if (!ok)
    return ok;

  /* If there are any SLP instances mark them as pure_slp.  */
  bool slp = vect_make_slp_decision (loop_vinfo);
  if (slp)
    {
      /* Find stmts that need to be both vectorized and SLPed.  */
      vect_detect_hybrid_slp (loop_vinfo);

      /* Update the vectorization factor based on the SLP decision.  */
      vect_update_vf_for_slp (loop_vinfo);

      /* Optimize the SLP graph with the vectorization factor fixed.  */
      vect_optimize_slp (loop_vinfo);
    }

  bool saved_can_fully_mask_p = LOOP_VINFO_CAN_FULLY_MASK_P (loop_vinfo);

  /* We don't expect to have to roll back to anything other than an empty
     set of rgroups.  */
  gcc_assert (LOOP_VINFO_MASKS (loop_vinfo).is_empty ());

  /* This is the point where we can re-start analysis with SLP forced off.  */
start_over:

  /* Now the vectorization factor is final.  */
  poly_uint64 vectorization_factor = LOOP_VINFO_VECT_FACTOR (loop_vinfo);
  gcc_assert (known_ne (vectorization_factor, 0U));

  if (LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo) && dump_enabled_p ())
    {
      dump_printf_loc (MSG_NOTE, vect_location,
		       "vectorization_factor = ");
      dump_dec (MSG_NOTE, vectorization_factor);
      dump_printf (MSG_NOTE, ", niters = %wd\n",
		   LOOP_VINFO_INT_NITERS (loop_vinfo));
    }

  /* Analyze the alignment of the data-refs in the loop.
     Fail if a data reference is found that cannot be vectorized.  */

  ok = vect_analyze_data_refs_alignment (loop_vinfo);
  if (!ok)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "bad data alignment.\n");
      return ok;
    }

  /* Prune the list of ddrs to be tested at run-time by versioning for alias.
     It is important to call pruning after vect_analyze_data_ref_accesses,
     since we use grouping information gathered by interleaving analysis.  */
  ok = vect_prune_runtime_alias_test_list (loop_vinfo);
  if (!ok)
    return ok;

  /* Do not invoke vect_enhance_data_refs_alignment for epilogue
     vectorization, since we do not want to add extra peeling or
     add versioning for alignment.  */
  if (!LOOP_VINFO_EPILOGUE_P (loop_vinfo))
    /* This pass will decide on using loop versioning and/or loop peeling in
       order to enhance the alignment of data references in the loop.  */
    ok = vect_enhance_data_refs_alignment (loop_vinfo);
  else
    ok = vect_verify_datarefs_alignment (loop_vinfo);
  if (!ok)
    return ok;

  if (slp)
    {
      /* Analyze operations in the SLP instances.  Note this may
	 remove unsupported SLP instances which makes the above
	 SLP kind detection invalid.  */
      unsigned old_size = LOOP_VINFO_SLP_INSTANCES (loop_vinfo).length ();
      vect_slp_analyze_operations (loop_vinfo);
      if (LOOP_VINFO_SLP_INSTANCES (loop_vinfo).length () != old_size)
	{
	  ok = opt_result::failure_at (vect_location,
				       "unsupported SLP instances\n");
	  goto again;
	}
    }

  /* Dissolve SLP-only groups.  */
  vect_dissolve_slp_only_groups (loop_vinfo);

  /* Scan all the remaining operations in the loop that are not subject
     to SLP and make sure they are vectorizable.  */
  ok = vect_analyze_loop_operations (loop_vinfo);
  if (!ok)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "bad operation or unsupported loop bound.\n");
      return ok;
    }

  /* Decide whether to use a fully-masked loop for this vectorization
     factor.  */
  LOOP_VINFO_FULLY_MASKED_P (loop_vinfo)
    = (LOOP_VINFO_CAN_FULLY_MASK_P (loop_vinfo)
       && vect_verify_full_masking (loop_vinfo));
  if (dump_enabled_p ())
    {
      if (LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
	dump_printf_loc (MSG_NOTE, vect_location,
			 "using a fully-masked loop.\n");
      else
	dump_printf_loc (MSG_NOTE, vect_location,
			 "not using a fully-masked loop.\n");
    }

  /* If epilog loop is required because of data accesses with gaps,
     one additional iteration needs to be peeled.  Check if there is
     enough iterations for vectorization.  */
  if (LOOP_VINFO_PEELING_FOR_GAPS (loop_vinfo)
      && LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo)
      && !LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
    {
      poly_uint64 vf = LOOP_VINFO_VECT_FACTOR (loop_vinfo);
      tree scalar_niters = LOOP_VINFO_NITERSM1 (loop_vinfo);

      if (known_lt (wi::to_widest (scalar_niters), vf))
	return opt_result::failure_at (vect_location,
				       "loop has no enough iterations to"
				       " support peeling for gaps.\n");
    }

  /* If we're vectorizing an epilogue loop, we either need a fully-masked
     loop or a loop that has a lower VF than the main loop.  */
  if (LOOP_VINFO_EPILOGUE_P (loop_vinfo)
      && !LOOP_VINFO_FULLY_MASKED_P (loop_vinfo)
      && maybe_ge (LOOP_VINFO_VECT_FACTOR (loop_vinfo),
		   LOOP_VINFO_VECT_FACTOR (orig_loop_vinfo)))
    return opt_result::failure_at (vect_location,
				   "Vectorization factor too high for"
				   " epilogue loop.\n");

  /* Check the costings of the loop make vectorizing worthwhile.  */
  res = vect_analyze_loop_costing (loop_vinfo);
  if (res < 0)
    {
      ok = opt_result::failure_at (vect_location,
				   "Loop costings may not be worthwhile.\n");
      goto again;
    }
  if (!res)
    return opt_result::failure_at (vect_location,
				   "Loop costings not worthwhile.\n");

  determine_peel_for_niter (loop_vinfo);
  /* If an epilogue loop is required make sure we can create one.  */
  if (LOOP_VINFO_PEELING_FOR_GAPS (loop_vinfo)
      || LOOP_VINFO_PEELING_FOR_NITER (loop_vinfo))
    {
      if (dump_enabled_p ())
        dump_printf_loc (MSG_NOTE, vect_location, "epilog loop required\n");
      if (!vect_can_advance_ivs_p (loop_vinfo)
	  || !slpeel_can_duplicate_loop_p (LOOP_VINFO_LOOP (loop_vinfo),
					   single_exit (LOOP_VINFO_LOOP
							 (loop_vinfo))))
        {
	  ok = opt_result::failure_at (vect_location,
				       "not vectorized: can't create required "
				       "epilog loop\n");
          goto again;
        }
    }

  /* During peeling, we need to check if number of loop iterations is
     enough for both peeled prolog loop and vector loop.  This check
     can be merged along with threshold check of loop versioning, so
     increase threshold for this case if necessary.

     If we are analyzing an epilogue we still want to check what its
     versioning threshold would be.  If we decide to vectorize the epilogues we
     will want to use the lowest versioning threshold of all epilogues and main
     loop.  This will enable us to enter a vectorized epilogue even when
     versioning the loop.  We can't simply check whether the epilogue requires
     versioning though since we may have skipped some versioning checks when
     analyzing the epilogue.  For instance, checks for alias versioning will be
     skipped when dealing with epilogues as we assume we already checked them
     for the main loop.  So instead we always check the 'orig_loop_vinfo'.  */
  if (LOOP_REQUIRES_VERSIONING (orig_loop_vinfo))
    {
      poly_uint64 niters_th = 0;
      unsigned int th = LOOP_VINFO_COST_MODEL_THRESHOLD (loop_vinfo);

      if (!vect_use_loop_mask_for_alignment_p (loop_vinfo))
	{
	  /* Niters for peeled prolog loop.  */
	  if (LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo) < 0)
	    {
	      dr_vec_info *dr_info = LOOP_VINFO_UNALIGNED_DR (loop_vinfo);
	      tree vectype = STMT_VINFO_VECTYPE (dr_info->stmt);
	      niters_th += TYPE_VECTOR_SUBPARTS (vectype) - 1;
	    }
	  else
	    niters_th += LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo);
	}

      /* Niters for at least one iteration of vectorized loop.  */
      if (!LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
	niters_th += LOOP_VINFO_VECT_FACTOR (loop_vinfo);
      /* One additional iteration because of peeling for gap.  */
      if (LOOP_VINFO_PEELING_FOR_GAPS (loop_vinfo))
	niters_th += 1;

      /*  Use the same condition as vect_transform_loop to decide when to use
	  the cost to determine a versioning threshold.  */
      if (vect_apply_runtime_profitability_check_p (loop_vinfo)
	  && ordered_p (th, niters_th))
	niters_th = ordered_max (poly_uint64 (th), niters_th);

      LOOP_VINFO_VERSIONING_THRESHOLD (loop_vinfo) = niters_th;
    }

  gcc_assert (known_eq (vectorization_factor,
			LOOP_VINFO_VECT_FACTOR (loop_vinfo)));

  /* Ok to vectorize!  */
  return opt_result::success ();

again:
  /* Ensure that "ok" is false (with an opt_problem if dumping is enabled).  */
  gcc_assert (!ok);

  /* Try again with SLP forced off but if we didn't do any SLP there is
     no point in re-trying.  */
  if (!slp)
    return ok;

  /* If there are reduction chains re-trying will fail anyway.  */
  if (! LOOP_VINFO_REDUCTION_CHAINS (loop_vinfo).is_empty ())
    return ok;

  /* Likewise if the grouped loads or stores in the SLP cannot be handled
     via interleaving or lane instructions.  */
  slp_instance instance;
  slp_tree node;
  unsigned i, j;
  FOR_EACH_VEC_ELT (LOOP_VINFO_SLP_INSTANCES (loop_vinfo), i, instance)
    {
      stmt_vec_info vinfo;
      vinfo = SLP_TREE_SCALAR_STMTS (SLP_INSTANCE_TREE (instance))[0];
      if (! STMT_VINFO_GROUPED_ACCESS (vinfo))
	continue;
      vinfo = DR_GROUP_FIRST_ELEMENT (vinfo);
      unsigned int size = DR_GROUP_SIZE (vinfo);
      tree vectype = STMT_VINFO_VECTYPE (vinfo);
      if (! vect_store_lanes_supported (vectype, size, false)
	 && ! known_eq (TYPE_VECTOR_SUBPARTS (vectype), 1U)
	 && ! vect_grouped_store_supported (vectype, size))
	return opt_result::failure_at (vinfo->stmt,
				       "unsupported grouped store\n");
      FOR_EACH_VEC_ELT (SLP_INSTANCE_LOADS (instance), j, node)
	{
	  vinfo = SLP_TREE_SCALAR_STMTS (node)[0];
	  vinfo = DR_GROUP_FIRST_ELEMENT (vinfo);
	  bool single_element_p = !DR_GROUP_NEXT_ELEMENT (vinfo);
	  size = DR_GROUP_SIZE (vinfo);
	  vectype = STMT_VINFO_VECTYPE (vinfo);
	  if (! vect_load_lanes_supported (vectype, size, false)
	      && ! vect_grouped_load_supported (vectype, single_element_p,
						size))
	    return opt_result::failure_at (vinfo->stmt,
					   "unsupported grouped load\n");
	}
    }

  if (dump_enabled_p ())
    dump_printf_loc (MSG_NOTE, vect_location,
		     "re-trying with SLP disabled\n");

  /* Roll back state appropriately.  No SLP this time.  */
  slp = false;
  /* Restore vectorization factor as it were without SLP.  */
  LOOP_VINFO_VECT_FACTOR (loop_vinfo) = saved_vectorization_factor;
  /* Free the SLP instances.  */
  FOR_EACH_VEC_ELT (LOOP_VINFO_SLP_INSTANCES (loop_vinfo), j, instance)
    vect_free_slp_instance (instance, false);
  LOOP_VINFO_SLP_INSTANCES (loop_vinfo).release ();
  /* Reset SLP type to loop_vect on all stmts.  */
  for (i = 0; i < LOOP_VINFO_LOOP (loop_vinfo)->num_nodes; ++i)
    {
      basic_block bb = LOOP_VINFO_BBS (loop_vinfo)[i];
      for (gimple_stmt_iterator si = gsi_start_phis (bb);
	   !gsi_end_p (si); gsi_next (&si))
	{
	  stmt_vec_info stmt_info = loop_vinfo->lookup_stmt (gsi_stmt (si));
	  STMT_SLP_TYPE (stmt_info) = loop_vect;
	  if (STMT_VINFO_DEF_TYPE (stmt_info) == vect_reduction_def
	      || STMT_VINFO_DEF_TYPE (stmt_info) == vect_double_reduction_def)
	    {
	      /* vectorizable_reduction adjusts reduction stmt def-types,
		 restore them to that of the PHI.  */
	      STMT_VINFO_DEF_TYPE (STMT_VINFO_REDUC_DEF (stmt_info))
		= STMT_VINFO_DEF_TYPE (stmt_info);
	      STMT_VINFO_DEF_TYPE (vect_stmt_to_vectorize
					(STMT_VINFO_REDUC_DEF (stmt_info)))
		= STMT_VINFO_DEF_TYPE (stmt_info);
	    }
	}
      for (gimple_stmt_iterator si = gsi_start_bb (bb);
	   !gsi_end_p (si); gsi_next (&si))
	{
	  stmt_vec_info stmt_info = loop_vinfo->lookup_stmt (gsi_stmt (si));
	  STMT_SLP_TYPE (stmt_info) = loop_vect;
	  if (STMT_VINFO_IN_PATTERN_P (stmt_info))
	    {
	      gimple *pattern_def_seq = STMT_VINFO_PATTERN_DEF_SEQ (stmt_info);
	      stmt_info = STMT_VINFO_RELATED_STMT (stmt_info);
	      STMT_SLP_TYPE (stmt_info) = loop_vect;
	      for (gimple_stmt_iterator pi = gsi_start (pattern_def_seq);
		   !gsi_end_p (pi); gsi_next (&pi))
		STMT_SLP_TYPE (loop_vinfo->lookup_stmt (gsi_stmt (pi)))
		  = loop_vect;
	    }
	}
    }
  /* Free optimized alias test DDRS.  */
  LOOP_VINFO_LOWER_BOUNDS (loop_vinfo).truncate (0);
  LOOP_VINFO_COMP_ALIAS_DDRS (loop_vinfo).release ();
  LOOP_VINFO_CHECK_UNEQUAL_ADDRS (loop_vinfo).release ();
  /* Reset target cost data.  */
  destroy_cost_data (LOOP_VINFO_TARGET_COST_DATA (loop_vinfo));
  LOOP_VINFO_TARGET_COST_DATA (loop_vinfo)
    = init_cost (LOOP_VINFO_LOOP (loop_vinfo));
  /* Reset accumulated rgroup information.  */
  release_vec_loop_masks (&LOOP_VINFO_MASKS (loop_vinfo));
  /* Reset assorted flags.  */
  LOOP_VINFO_PEELING_FOR_NITER (loop_vinfo) = false;
  LOOP_VINFO_PEELING_FOR_GAPS (loop_vinfo) = false;
  LOOP_VINFO_COST_MODEL_THRESHOLD (loop_vinfo) = 0;
  LOOP_VINFO_VERSIONING_THRESHOLD (loop_vinfo) = 0;
  LOOP_VINFO_CAN_FULLY_MASK_P (loop_vinfo) = saved_can_fully_mask_p;

  goto start_over;
}

/* Return true if vectorizing a loop using NEW_LOOP_VINFO appears
   to be better than vectorizing it using OLD_LOOP_VINFO.  Assume that
   OLD_LOOP_VINFO is better unless something specifically indicates
   otherwise.

   Note that this deliberately isn't a partial order.  */

static bool
vect_better_loop_vinfo_p (loop_vec_info new_loop_vinfo,
			  loop_vec_info old_loop_vinfo)
{
  struct loop *loop = LOOP_VINFO_LOOP (new_loop_vinfo);
  gcc_assert (LOOP_VINFO_LOOP (old_loop_vinfo) == loop);

  poly_int64 new_vf = LOOP_VINFO_VECT_FACTOR (new_loop_vinfo);
  poly_int64 old_vf = LOOP_VINFO_VECT_FACTOR (old_loop_vinfo);

  /* Always prefer a VF of loop->simdlen over any other VF.  */
  if (loop->simdlen)
    {
      bool new_simdlen_p = known_eq (new_vf, loop->simdlen);
      bool old_simdlen_p = known_eq (old_vf, loop->simdlen);
      if (new_simdlen_p != old_simdlen_p)
	return new_simdlen_p;
    }

  /* Limit the VFs to what is likely to be the maximum number of iterations,
     to handle cases in which at least one loop_vinfo is fully-masked.  */
  HOST_WIDE_INT estimated_max_niter = likely_max_stmt_executions_int (loop);
  if (estimated_max_niter != -1)
    {
      if (known_le (estimated_max_niter, new_vf))
	new_vf = estimated_max_niter;
      if (known_le (estimated_max_niter, old_vf))
	old_vf = estimated_max_niter;
    }

  /* Check whether the (fractional) cost per scalar iteration is lower
     or higher: new_inside_cost / new_vf vs. old_inside_cost / old_vf.  */
  poly_widest_int rel_new = (new_loop_vinfo->vec_inside_cost
			     * poly_widest_int (old_vf));
  poly_widest_int rel_old = (old_loop_vinfo->vec_inside_cost
			     * poly_widest_int (new_vf));
  if (maybe_lt (rel_old, rel_new))
    {
      /* When old_loop_vinfo uses a variable vectorization factor,
	 we know that it has a lower cost for at least one runtime VF.
	 However, we don't know how likely that VF is.

	 One option would be to compare the costs for the estimated VFs.
	 The problem is that that can put too much pressure on the cost
	 model.  E.g. if the estimated VF is also the lowest possible VF,
	 and if old_loop_vinfo is 1 unit worse than new_loop_vinfo
	 for the estimated VF, we'd then choose new_loop_vinfo even
	 though (a) new_loop_vinfo might not actually be better than
	 old_loop_vinfo for that VF and (b) it would be significantly
	 worse at larger VFs.

	 Here we go for a hacky compromise: pick new_loop_vinfo if it is
	 no more expensive than old_loop_vinfo even after doubling the
	 estimated old_loop_vinfo VF.  For all but trivial loops, this
	 ensures that we only pick new_loop_vinfo if it is significantly
	 better than old_loop_vinfo at the estimated VF.  */
      if (rel_new.is_constant ())
	return false;

      HOST_WIDE_INT new_estimated_vf = estimated_poly_value (new_vf);
      HOST_WIDE_INT old_estimated_vf = estimated_poly_value (old_vf);
      widest_int estimated_rel_new = (new_loop_vinfo->vec_inside_cost
				      * widest_int (old_estimated_vf));
      widest_int estimated_rel_old = (old_loop_vinfo->vec_inside_cost
				      * widest_int (new_estimated_vf));
      return estimated_rel_new * 2 <= estimated_rel_old;
    }
  if (known_lt (rel_new, rel_old))
    return true;

  /* If there's nothing to choose between the loop bodies, see whether
     there's a difference in the prologue and epilogue costs.  */
  if (new_loop_vinfo->vec_outside_cost != old_loop_vinfo->vec_outside_cost)
    return new_loop_vinfo->vec_outside_cost < old_loop_vinfo->vec_outside_cost;

  return false;
}

/* Decide whether to replace OLD_LOOP_VINFO with NEW_LOOP_VINFO.  Return
   true if we should.  */

static bool
vect_joust_loop_vinfos (loop_vec_info new_loop_vinfo,
			loop_vec_info old_loop_vinfo)
{
  if (!vect_better_loop_vinfo_p (new_loop_vinfo, old_loop_vinfo))
    return false;

  if (dump_enabled_p ())
    dump_printf_loc (MSG_NOTE, vect_location,
		     "***** Preferring vector mode %s to vector mode %s\n",
		     GET_MODE_NAME (new_loop_vinfo->vector_mode),
		     GET_MODE_NAME (old_loop_vinfo->vector_mode));
  return true;
}

/* Function vect_analyze_loop.

   Apply a set of analyses on LOOP, and create a loop_vec_info struct
   for it.  The different analyses will record information in the
   loop_vec_info struct.  */
opt_loop_vec_info
vect_analyze_loop (class loop *loop, vec_info_shared *shared)
{
  auto_vector_modes vector_modes;

  /* Autodetect first vector size we try.  */
  unsigned int autovec_flags
    = targetm.vectorize.autovectorize_vector_modes (&vector_modes,
						    loop->simdlen != 0);
  unsigned int mode_i = 0;

  DUMP_VECT_SCOPE ("analyze_loop_nest");

  if (loop_outer (loop)
      && loop_vec_info_for_loop (loop_outer (loop))
      && LOOP_VINFO_VECTORIZABLE_P (loop_vec_info_for_loop (loop_outer (loop))))
    return opt_loop_vec_info::failure_at (vect_location,
					  "outer-loop already vectorized.\n");

  if (!find_loop_nest (loop, &shared->loop_nest))
    return opt_loop_vec_info::failure_at
      (vect_location,
       "not vectorized: loop nest containing two or more consecutive inner"
       " loops cannot be vectorized\n");

  unsigned n_stmts = 0;
  machine_mode autodetected_vector_mode = VOIDmode;
  opt_loop_vec_info first_loop_vinfo = opt_loop_vec_info::success (NULL);
  machine_mode next_vector_mode = VOIDmode;
  poly_uint64 lowest_th = 0;
  unsigned vectorized_loops = 0;
  bool pick_lowest_cost_p = ((autovec_flags & VECT_COMPARE_COSTS)
			     && !unlimited_cost_model (loop));

  bool vect_epilogues = false;
  opt_result res = opt_result::success ();
  unsigned HOST_WIDE_INT simdlen = loop->simdlen;
  while (1)
    {
      /* Check the CFG characteristics of the loop (nesting, entry/exit).  */
      opt_loop_vec_info loop_vinfo = vect_analyze_loop_form (loop, shared);
      if (!loop_vinfo)
	{
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			     "bad loop form.\n");
	  gcc_checking_assert (first_loop_vinfo == NULL);
	  return loop_vinfo;
	}
      loop_vinfo->vector_mode = next_vector_mode;

      bool fatal = false;

      /* When pick_lowest_cost_p is true, we should in principle iterate
	 over all the loop_vec_infos that LOOP_VINFO could replace and
	 try to vectorize LOOP_VINFO under the same conditions.
	 E.g. when trying to replace an epilogue loop, we should vectorize
	 LOOP_VINFO as an epilogue loop with the same VF limit.  When trying
	 to replace the main loop, we should vectorize LOOP_VINFO as a main
	 loop too.

	 However, autovectorize_vector_modes is usually sorted as follows:

	 - Modes that naturally produce lower VFs usually follow modes that
	   naturally produce higher VFs.

	 - When modes naturally produce the same VF, maskable modes
	   usually follow unmaskable ones, so that the maskable mode
	   can be used to vectorize the epilogue of the unmaskable mode.

	 This order is preferred because it leads to the maximum
	 epilogue vectorization opportunities.  Targets should only use
	 a different order if they want to make wide modes available while
	 disparaging them relative to earlier, smaller modes.  The assumption
	 in that case is that the wider modes are more expensive in some
	 way that isn't reflected directly in the costs.

	 There should therefore be few interesting cases in which
	 LOOP_VINFO fails when treated as an epilogue loop, succeeds when
	 treated as a standalone loop, and ends up being genuinely cheaper
	 than FIRST_LOOP_VINFO.  */
      if (vect_epilogues)
	LOOP_VINFO_ORIG_LOOP_INFO (loop_vinfo) = first_loop_vinfo;

      res = vect_analyze_loop_2 (loop_vinfo, fatal, &n_stmts);
      if (mode_i == 0)
	autodetected_vector_mode = loop_vinfo->vector_mode;
      if (dump_enabled_p ())
	{
	  if (res)
	    dump_printf_loc (MSG_NOTE, vect_location,
			     "***** Analysis succeeded with vector mode %s\n",
			     GET_MODE_NAME (loop_vinfo->vector_mode));
	  else
	    dump_printf_loc (MSG_NOTE, vect_location,
			     "***** Analysis failed with vector mode %s\n",
			     GET_MODE_NAME (loop_vinfo->vector_mode));
	}

      loop->aux = NULL;

      if (!fatal)
	while (mode_i < vector_modes.length ()
	       && vect_chooses_same_modes_p (loop_vinfo, vector_modes[mode_i]))
	  {
	    if (dump_enabled_p ())
	      dump_printf_loc (MSG_NOTE, vect_location,
			       "***** The result for vector mode %s would"
			       " be the same\n",
			       GET_MODE_NAME (vector_modes[mode_i]));
	    mode_i += 1;
	  }

      if (res)
	{
	  LOOP_VINFO_VECTORIZABLE_P (loop_vinfo) = 1;
	  vectorized_loops++;

	  /* Once we hit the desired simdlen for the first time,
	     discard any previous attempts.  */
	  if (simdlen
	      && known_eq (LOOP_VINFO_VECT_FACTOR (loop_vinfo), simdlen))
	    {
	      delete first_loop_vinfo;
	      first_loop_vinfo = opt_loop_vec_info::success (NULL);
	      LOOP_VINFO_ORIG_LOOP_INFO (loop_vinfo) = NULL;
	      simdlen = 0;
	    }
	  else if (pick_lowest_cost_p && first_loop_vinfo)
	    {
	      /* Keep trying to roll back vectorization attempts while the
		 loop_vec_infos they produced were worse than this one.  */
	      vec<loop_vec_info> &vinfos = first_loop_vinfo->epilogue_vinfos;
	      while (!vinfos.is_empty ()
		     && vect_joust_loop_vinfos (loop_vinfo, vinfos.last ()))
		{
		  gcc_assert (vect_epilogues);
		  delete vinfos.pop ();
		}
	      if (vinfos.is_empty ()
		  && vect_joust_loop_vinfos (loop_vinfo, first_loop_vinfo))
		{
		  delete first_loop_vinfo;
		  first_loop_vinfo = opt_loop_vec_info::success (NULL);
		  LOOP_VINFO_ORIG_LOOP_INFO (loop_vinfo) = NULL;
		}
	    }

	  if (first_loop_vinfo == NULL)
	    {
	      first_loop_vinfo = loop_vinfo;
	      lowest_th = LOOP_VINFO_VERSIONING_THRESHOLD (first_loop_vinfo);
	    }
	  else if (vect_epilogues
		   /* For now only allow one epilogue loop.  */
		   && first_loop_vinfo->epilogue_vinfos.is_empty ())
	    {
	      first_loop_vinfo->epilogue_vinfos.safe_push (loop_vinfo);
	      poly_uint64 th = LOOP_VINFO_VERSIONING_THRESHOLD (loop_vinfo);
	      gcc_assert (!LOOP_REQUIRES_VERSIONING (loop_vinfo)
			  || maybe_ne (lowest_th, 0U));
	      /* Keep track of the known smallest versioning
		 threshold.  */
	      if (ordered_p (lowest_th, th))
		lowest_th = ordered_min (lowest_th, th);
	    }
	  else
	    delete loop_vinfo;

	  /* Only vectorize epilogues if PARAM_VECT_EPILOGUES_NOMASK is
	     enabled, SIMDUID is not set, it is the innermost loop and we have
	     either already found the loop's SIMDLEN or there was no SIMDLEN to
	     begin with.
	     TODO: Enable epilogue vectorization for loops with SIMDUID set.  */
	  vect_epilogues = (!simdlen
			    && loop->inner == NULL
			    && param_vect_epilogues_nomask
			    && LOOP_VINFO_PEELING_FOR_NITER (first_loop_vinfo)
			    && !loop->simduid
			    /* For now only allow one epilogue loop, but allow
			       pick_lowest_cost_p to replace it.  */
			    && (first_loop_vinfo->epilogue_vinfos.is_empty ()
				|| pick_lowest_cost_p));

	  /* Commit to first_loop_vinfo if we have no reason to try
	     alternatives.  */
	  if (!simdlen && !vect_epilogues && !pick_lowest_cost_p)
	    break;
	}
      else
	{
	  delete loop_vinfo;
	  if (fatal)
	    {
	      gcc_checking_assert (first_loop_vinfo == NULL);
	      break;
	    }
	}

      if (mode_i < vector_modes.length ()
	  && VECTOR_MODE_P (autodetected_vector_mode)
	  && (related_vector_mode (vector_modes[mode_i],
				   GET_MODE_INNER (autodetected_vector_mode))
	      == autodetected_vector_mode)
	  && (related_vector_mode (autodetected_vector_mode,
				   GET_MODE_INNER (vector_modes[mode_i]))
	      == vector_modes[mode_i]))
	{
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_NOTE, vect_location,
			     "***** Skipping vector mode %s, which would"
			     " repeat the analysis for %s\n",
			     GET_MODE_NAME (vector_modes[mode_i]),
			     GET_MODE_NAME (autodetected_vector_mode));
	  mode_i += 1;
	}

      if (mode_i == vector_modes.length ()
	  || autodetected_vector_mode == VOIDmode)
	break;

      /* Try the next biggest vector size.  */
      next_vector_mode = vector_modes[mode_i++];
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location,
			 "***** Re-trying analysis with vector mode %s\n",
			 GET_MODE_NAME (next_vector_mode));
    }

  if (first_loop_vinfo)
    {
      loop->aux = (loop_vec_info) first_loop_vinfo;
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location,
			 "***** Choosing vector mode %s\n",
			 GET_MODE_NAME (first_loop_vinfo->vector_mode));
      LOOP_VINFO_VERSIONING_THRESHOLD (first_loop_vinfo) = lowest_th;
      return first_loop_vinfo;
    }

  return opt_loop_vec_info::propagate_failure (res);
}

/* Return true if there is an in-order reduction function for CODE, storing
   it in *REDUC_FN if so.  */

static bool
fold_left_reduction_fn (tree_code code, internal_fn *reduc_fn)
{
  switch (code)
    {
    case PLUS_EXPR:
      *reduc_fn = IFN_FOLD_LEFT_PLUS;
      return true;

    default:
      return false;
    }
}

/* Function reduction_fn_for_scalar_code

   Input:
   CODE - tree_code of a reduction operations.

   Output:
   REDUC_FN - the corresponding internal function to be used to reduce the
      vector of partial results into a single scalar result, or IFN_LAST
      if the operation is a supported reduction operation, but does not have
      such an internal function.

   Return FALSE if CODE currently cannot be vectorized as reduction.  */

static bool
reduction_fn_for_scalar_code (enum tree_code code, internal_fn *reduc_fn)
{
  switch (code)
    {
      case MAX_EXPR:
        *reduc_fn = IFN_REDUC_MAX;
        return true;

      case MIN_EXPR:
        *reduc_fn = IFN_REDUC_MIN;
        return true;

      case PLUS_EXPR:
        *reduc_fn = IFN_REDUC_PLUS;
        return true;

      case BIT_AND_EXPR:
	*reduc_fn = IFN_REDUC_AND;
	return true;

      case BIT_IOR_EXPR:
	*reduc_fn = IFN_REDUC_IOR;
	return true;

      case BIT_XOR_EXPR:
	*reduc_fn = IFN_REDUC_XOR;
	return true;

      case MULT_EXPR:
      case MINUS_EXPR:
        *reduc_fn = IFN_LAST;
        return true;

      default:
       return false;
    }
}

/* If there is a neutral value X such that SLP reduction NODE would not
   be affected by the introduction of additional X elements, return that X,
   otherwise return null.  CODE is the code of the reduction and VECTOR_TYPE
   is the vector type that would hold element X.  REDUC_CHAIN is true if
   the SLP statements perform a single reduction, false if each statement
   performs an independent reduction.  */

static tree
neutral_op_for_slp_reduction (slp_tree slp_node, tree vector_type,
			      tree_code code, bool reduc_chain)
{
  vec<stmt_vec_info> stmts = SLP_TREE_SCALAR_STMTS (slp_node);
  stmt_vec_info stmt_vinfo = stmts[0];
  tree scalar_type = TREE_TYPE (vector_type);
  class loop *loop = gimple_bb (stmt_vinfo->stmt)->loop_father;
  gcc_assert (loop);

  switch (code)
    {
    case WIDEN_SUM_EXPR:
    case DOT_PROD_EXPR:
    case SAD_EXPR:
    case PLUS_EXPR:
    case MINUS_EXPR:
    case BIT_IOR_EXPR:
    case BIT_XOR_EXPR:
      return build_zero_cst (scalar_type);

    case MULT_EXPR:
      return build_one_cst (scalar_type);

    case BIT_AND_EXPR:
      return build_all_ones_cst (scalar_type);

    case MAX_EXPR:
    case MIN_EXPR:
      /* For MIN/MAX the initial values are neutral.  A reduction chain
	 has only a single initial value, so that value is neutral for
	 all statements.  */
      if (reduc_chain)
	return PHI_ARG_DEF_FROM_EDGE (stmt_vinfo->stmt,
				      loop_preheader_edge (loop));
      return NULL_TREE;

    default:
      return NULL_TREE;
    }
}

/* Error reporting helper for vect_is_simple_reduction below.  GIMPLE statement
   STMT is printed with a message MSG. */

static void
report_vect_op (dump_flags_t msg_type, gimple *stmt, const char *msg)
{
  dump_printf_loc (msg_type, vect_location, "%s%G", msg, stmt);
}

/* Return true if we need an in-order reduction for operation CODE
   on type TYPE.  NEED_WRAPPING_INTEGRAL_OVERFLOW is true if integer
   overflow must wrap.  */

bool
needs_fold_left_reduction_p (tree type, tree_code code)
{
  /* CHECKME: check for !flag_finite_math_only too?  */
  if (SCALAR_FLOAT_TYPE_P (type))
    switch (code)
      {
      case MIN_EXPR:
      case MAX_EXPR:
	return false;

      default:
	return !flag_associative_math;
      }

  if (INTEGRAL_TYPE_P (type))
    {
      if (!operation_no_trapping_overflow (type, code))
	return true;
      return false;
    }

  if (SAT_FIXED_POINT_TYPE_P (type))
    return true;

  return false;
}

/* Return true if the reduction PHI in LOOP with latch arg LOOP_ARG and
   has a handled computation expression.  Store the main reduction
   operation in *CODE.  */

static bool
check_reduction_path (dump_user_location_t loc, loop_p loop, gphi *phi,
		      tree loop_arg, enum tree_code *code,
		      vec<std::pair<ssa_op_iter, use_operand_p> > &path)
{
  auto_bitmap visited;
  tree lookfor = PHI_RESULT (phi);
  ssa_op_iter curri;
  use_operand_p curr = op_iter_init_phiuse (&curri, phi, SSA_OP_USE);
  while (USE_FROM_PTR (curr) != loop_arg)
    curr = op_iter_next_use (&curri);
  curri.i = curri.numops;
  do
    {
      path.safe_push (std::make_pair (curri, curr));
      tree use = USE_FROM_PTR (curr);
      if (use == lookfor)
	break;
      gimple *def = SSA_NAME_DEF_STMT (use);
      if (gimple_nop_p (def)
	  || ! flow_bb_inside_loop_p (loop, gimple_bb (def)))
	{
pop:
	  do
	    {
	      std::pair<ssa_op_iter, use_operand_p> x = path.pop ();
	      curri = x.first;
	      curr = x.second;
	      do
		curr = op_iter_next_use (&curri);
	      /* Skip already visited or non-SSA operands (from iterating
	         over PHI args).  */
	      while (curr != NULL_USE_OPERAND_P
		     && (TREE_CODE (USE_FROM_PTR (curr)) != SSA_NAME
			 || ! bitmap_set_bit (visited,
					      SSA_NAME_VERSION
					        (USE_FROM_PTR (curr)))));
	    }
	  while (curr == NULL_USE_OPERAND_P && ! path.is_empty ());
	  if (curr == NULL_USE_OPERAND_P)
	    break;
	}
      else
	{
	  if (gimple_code (def) == GIMPLE_PHI)
	    curr = op_iter_init_phiuse (&curri, as_a <gphi *>(def), SSA_OP_USE);
	  else
	    curr = op_iter_init_use (&curri, def, SSA_OP_USE);
	  while (curr != NULL_USE_OPERAND_P
		 && (TREE_CODE (USE_FROM_PTR (curr)) != SSA_NAME
		     || ! bitmap_set_bit (visited,
					  SSA_NAME_VERSION
					    (USE_FROM_PTR (curr)))))
	    curr = op_iter_next_use (&curri);
	  if (curr == NULL_USE_OPERAND_P)
	    goto pop;
	}
    }
  while (1);
  if (dump_file && (dump_flags & TDF_DETAILS))
    {
      dump_printf_loc (MSG_NOTE, loc, "reduction path: ");
      unsigned i;
      std::pair<ssa_op_iter, use_operand_p> *x;
      FOR_EACH_VEC_ELT (path, i, x)
	dump_printf (MSG_NOTE, "%T ", USE_FROM_PTR (x->second));
      dump_printf (MSG_NOTE, "\n");
    }

  /* Check whether the reduction path detected is valid.  */
  bool fail = path.length () == 0;
  bool neg = false;
  int sign = -1;
  *code = ERROR_MARK;
  for (unsigned i = 1; i < path.length (); ++i)
    {
      gimple *use_stmt = USE_STMT (path[i].second);
      tree op = USE_FROM_PTR (path[i].second);
      if (! is_gimple_assign (use_stmt)
	  /* The following make sure we can compute the operand index
	     easily plus it mostly disallows chaining via COND_EXPR condition
	     operands.  */
	  || (gimple_assign_rhs1_ptr (use_stmt) != path[i].second->use
	      && (gimple_num_ops (use_stmt) <= 2
		  || gimple_assign_rhs2_ptr (use_stmt) != path[i].second->use)
	      && (gimple_num_ops (use_stmt) <= 3
		  || gimple_assign_rhs3_ptr (use_stmt) != path[i].second->use)))
	{
	  fail = true;
	  break;
	}
      /* Check there's only a single stmt the op is used on inside
         of the loop.  */
      imm_use_iterator imm_iter;
      gimple *op_use_stmt;
      unsigned cnt = 0;
      FOR_EACH_IMM_USE_STMT (op_use_stmt, imm_iter, op)
	if (!is_gimple_debug (op_use_stmt)
	    && flow_bb_inside_loop_p (loop, gimple_bb (op_use_stmt)))
	  {
	    /* We want to allow x + x but not x < 1 ? x : 2.  */
	    if (is_gimple_assign (op_use_stmt)
		&& gimple_assign_rhs_code (op_use_stmt) == COND_EXPR)
	      {
		use_operand_p use_p;
		FOR_EACH_IMM_USE_ON_STMT (use_p, imm_iter)
		  cnt++;
	      }
	    else
	      cnt++;
	  }
      if (cnt != 1)
	{
	  fail = true;
	  break;
	}
      tree_code use_code = gimple_assign_rhs_code (use_stmt);
      if (use_code == MINUS_EXPR)
	{
	  use_code = PLUS_EXPR;
	  /* Track whether we negate the reduction value each iteration.  */
	  if (gimple_assign_rhs2 (use_stmt) == op)
	    neg = ! neg;
	}
      if (CONVERT_EXPR_CODE_P (use_code)
	  && tree_nop_conversion_p (TREE_TYPE (gimple_assign_lhs (use_stmt)),
				    TREE_TYPE (gimple_assign_rhs1 (use_stmt))))
	;
      else if (*code == ERROR_MARK)
	{
	  *code = use_code;
	  sign = TYPE_SIGN (TREE_TYPE (gimple_assign_lhs (use_stmt)));
	}
      else if (use_code != *code)
	{
	  fail = true;
	  break;
	}
      else if ((use_code == MIN_EXPR
		|| use_code == MAX_EXPR)
	       && sign != TYPE_SIGN (TREE_TYPE (gimple_assign_lhs (use_stmt))))
	{
	  fail = true;
	  break;
	}
    }
  return ! fail && ! neg && *code != ERROR_MARK;
}

bool
check_reduction_path (dump_user_location_t loc, loop_p loop, gphi *phi,
		      tree loop_arg, enum tree_code code)
{
  auto_vec<std::pair<ssa_op_iter, use_operand_p> > path;
  enum tree_code code_;
  return (check_reduction_path (loc, loop, phi, loop_arg, &code_, path)
	  && code_ == code);
}



/* Function vect_is_simple_reduction

   (1) Detect a cross-iteration def-use cycle that represents a simple
   reduction computation.  We look for the following pattern:

   loop_header:
     a1 = phi < a0, a2 >
     a3 = ...
     a2 = operation (a3, a1)

   or

   a3 = ...
   loop_header:
     a1 = phi < a0, a2 >
     a2 = operation (a3, a1)

   such that:
   1. operation is commutative and associative and it is safe to
      change the order of the computation
   2. no uses for a2 in the loop (a2 is used out of the loop)
   3. no uses of a1 in the loop besides the reduction operation
   4. no uses of a1 outside the loop.

   Conditions 1,4 are tested here.
   Conditions 2,3 are tested in vect_mark_stmts_to_be_vectorized.

   (2) Detect a cross-iteration def-use cycle in nested loops, i.e.,
   nested cycles.

   (3) Detect cycles of phi nodes in outer-loop vectorization, i.e., double
   reductions:

     a1 = phi < a0, a2 >
     inner loop (def of a3)
     a2 = phi < a3 >

   (4) Detect condition expressions, ie:
     for (int i = 0; i < N; i++)
       if (a[i] < val)
	ret_val = a[i];

*/

static stmt_vec_info
vect_is_simple_reduction (loop_vec_info loop_info, stmt_vec_info phi_info,
			  bool *double_reduc, bool *reduc_chain_p)
{
  gphi *phi = as_a <gphi *> (phi_info->stmt);
  gimple *phi_use_stmt = NULL;
  imm_use_iterator imm_iter;
  use_operand_p use_p;

  *double_reduc = false;
  *reduc_chain_p = false;
  STMT_VINFO_REDUC_TYPE (phi_info) = TREE_CODE_REDUCTION;

  tree phi_name = PHI_RESULT (phi);
  /* ???  If there are no uses of the PHI result the inner loop reduction
     won't be detected as possibly double-reduction by vectorizable_reduction
     because that tries to walk the PHI arg from the preheader edge which
     can be constant.  See PR60382.  */
  if (has_zero_uses (phi_name))
    return NULL;
  class loop *loop = (gimple_bb (phi))->loop_father;
  unsigned nphi_def_loop_uses = 0;
  FOR_EACH_IMM_USE_FAST (use_p, imm_iter, phi_name)
    {
      gimple *use_stmt = USE_STMT (use_p);
      if (is_gimple_debug (use_stmt))
	continue;

      if (!flow_bb_inside_loop_p (loop, gimple_bb (use_stmt)))
        {
          if (dump_enabled_p ())
	    dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			     "intermediate value used outside loop.\n");

          return NULL;
        }

      nphi_def_loop_uses++;
      phi_use_stmt = use_stmt;
    }

  tree latch_def = PHI_ARG_DEF_FROM_EDGE (phi, loop_latch_edge (loop));
  if (TREE_CODE (latch_def) != SSA_NAME)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "reduction: not ssa_name: %T\n", latch_def);
      return NULL;
    }

  stmt_vec_info def_stmt_info = loop_info->lookup_def (latch_def);
  if (!def_stmt_info
      || !flow_bb_inside_loop_p (loop, gimple_bb (def_stmt_info->stmt)))
    return NULL;

  bool nested_in_vect_loop
    = flow_loop_nested_p (LOOP_VINFO_LOOP (loop_info), loop);
  unsigned nlatch_def_loop_uses = 0;
  auto_vec<gphi *, 3> lcphis;
  bool inner_loop_of_double_reduc = false;
  FOR_EACH_IMM_USE_FAST (use_p, imm_iter, latch_def)
    {
      gimple *use_stmt = USE_STMT (use_p);
      if (is_gimple_debug (use_stmt))
	continue;
      if (flow_bb_inside_loop_p (loop, gimple_bb (use_stmt)))
	nlatch_def_loop_uses++;
      else
	{
	  /* We can have more than one loop-closed PHI.  */
	  lcphis.safe_push (as_a <gphi *> (use_stmt));
	  if (nested_in_vect_loop
	      && (STMT_VINFO_DEF_TYPE (loop_info->lookup_stmt (use_stmt))
		  == vect_double_reduction_def))
	    inner_loop_of_double_reduc = true;
	}
    }

  /* If we are vectorizing an inner reduction we are executing that
     in the original order only in case we are not dealing with a
     double reduction.  */
  if (nested_in_vect_loop && !inner_loop_of_double_reduc)
    {
      if (dump_enabled_p ())
	report_vect_op (MSG_NOTE, def_stmt_info->stmt,
			"detected nested cycle: ");
      return def_stmt_info;
    }

  /* If this isn't a nested cycle or if the nested cycle reduction value
     is used ouside of the inner loop we cannot handle uses of the reduction
     value.  */
  if (nlatch_def_loop_uses > 1 || nphi_def_loop_uses > 1)
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "reduction used in loop.\n");
      return NULL;
    }

  /* If DEF_STMT is a phi node itself, we expect it to have a single argument
     defined in the inner loop.  */
  if (gphi *def_stmt = dyn_cast <gphi *> (def_stmt_info->stmt))
    {
      tree op1 = PHI_ARG_DEF (def_stmt, 0);
      if (gimple_phi_num_args (def_stmt) != 1
          || TREE_CODE (op1) != SSA_NAME)
        {
          if (dump_enabled_p ())
	    dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			     "unsupported phi node definition.\n");

          return NULL;
        }

      gimple *def1 = SSA_NAME_DEF_STMT (op1);
      if (gimple_bb (def1)
	  && flow_bb_inside_loop_p (loop, gimple_bb (def_stmt))
          && loop->inner
          && flow_bb_inside_loop_p (loop->inner, gimple_bb (def1))
          && is_gimple_assign (def1)
	  && is_a <gphi *> (phi_use_stmt)
	  && flow_bb_inside_loop_p (loop->inner, gimple_bb (phi_use_stmt)))
        {
          if (dump_enabled_p ())
            report_vect_op (MSG_NOTE, def_stmt,
			    "detected double reduction: ");

          *double_reduc = true;
	  return def_stmt_info;
        }

      return NULL;
    }

  /* Look for the expression computing latch_def from then loop PHI result.  */
  auto_vec<std::pair<ssa_op_iter, use_operand_p> > path;
  enum tree_code code;
  if (check_reduction_path (vect_location, loop, phi, latch_def, &code,
			    path))
    {
      STMT_VINFO_REDUC_CODE (phi_info) = code;
      if (code == COND_EXPR && !nested_in_vect_loop)
	STMT_VINFO_REDUC_TYPE (phi_info) = COND_REDUCTION;

      /* Fill in STMT_VINFO_REDUC_IDX and gather stmts for an SLP
	 reduction chain for which the additional restriction is that
	 all operations in the chain are the same.  */
      auto_vec<stmt_vec_info, 8> reduc_chain;
      unsigned i;
      bool is_slp_reduc = !nested_in_vect_loop && code != COND_EXPR;
      for (i = path.length () - 1; i >= 1; --i)
	{
	  gimple *stmt = USE_STMT (path[i].second);
	  stmt_vec_info stmt_info = loop_info->lookup_stmt (stmt);
	  STMT_VINFO_REDUC_IDX (stmt_info)
	    = path[i].second->use - gimple_assign_rhs1_ptr (stmt);
	  enum tree_code stmt_code = gimple_assign_rhs_code (stmt);
	  bool leading_conversion = (CONVERT_EXPR_CODE_P (stmt_code)
				     && (i == 1 || i == path.length () - 1));
	  if ((stmt_code != code && !leading_conversion)
	      /* We can only handle the final value in epilogue
		 generation for reduction chains.  */
	      || (i != 1 && !has_single_use (gimple_assign_lhs (stmt))))
	    is_slp_reduc = false;
	  /* For reduction chains we support a trailing/leading
	     conversions.  We do not store those in the actual chain.  */
	  if (leading_conversion)
	    continue;
	  reduc_chain.safe_push (stmt_info);
	}
      if (is_slp_reduc && reduc_chain.length () > 1)
	{
	  for (unsigned i = 0; i < reduc_chain.length () - 1; ++i)
	    {
	      REDUC_GROUP_FIRST_ELEMENT (reduc_chain[i]) = reduc_chain[0];
	      REDUC_GROUP_NEXT_ELEMENT (reduc_chain[i]) = reduc_chain[i+1];
	    }
	  REDUC_GROUP_FIRST_ELEMENT (reduc_chain.last ()) = reduc_chain[0];
	  REDUC_GROUP_NEXT_ELEMENT (reduc_chain.last ()) = NULL;

	  /* Save the chain for further analysis in SLP detection.  */
	  LOOP_VINFO_REDUCTION_CHAINS (loop_info).safe_push (reduc_chain[0]);
	  REDUC_GROUP_SIZE (reduc_chain[0]) = reduc_chain.length ();

	  *reduc_chain_p = true;
	  if (dump_enabled_p ())
	    dump_printf_loc (MSG_NOTE, vect_location,
			    "reduction: detected reduction chain\n");
	}
      else if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location,
			 "reduction: detected reduction\n");

      return def_stmt_info;
    }

  if (dump_enabled_p ())
    dump_printf_loc (MSG_NOTE, vect_location,
		     "reduction: unknown pattern\n");

  return NULL;
}

/* Calculate cost of peeling the loop PEEL_ITERS_PROLOGUE times.  */
int
vect_get_known_peeling_cost (loop_vec_info loop_vinfo, int peel_iters_prologue,
                             int *peel_iters_epilogue,
                             stmt_vector_for_cost *scalar_cost_vec,
			     stmt_vector_for_cost *prologue_cost_vec,
			     stmt_vector_for_cost *epilogue_cost_vec)
{
  int retval = 0;
  int assumed_vf = vect_vf_for_cost (loop_vinfo);

  if (!LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo))
    {
      *peel_iters_epilogue = assumed_vf / 2;
      if (dump_enabled_p ())
        dump_printf_loc (MSG_NOTE, vect_location,
			 "cost model: epilogue peel iters set to vf/2 "
			 "because loop iterations are unknown .\n");

      /* If peeled iterations are known but number of scalar loop
         iterations are unknown, count a taken branch per peeled loop.  */
      retval = record_stmt_cost (prologue_cost_vec, 1, cond_branch_taken,
				 NULL, NULL_TREE, 0, vect_prologue);
      retval += record_stmt_cost (epilogue_cost_vec, 1, cond_branch_taken,
				  NULL, NULL_TREE, 0, vect_epilogue);
    }
  else
    {
      int niters = LOOP_VINFO_INT_NITERS (loop_vinfo);
      peel_iters_prologue = niters < peel_iters_prologue ?
                            niters : peel_iters_prologue;
      *peel_iters_epilogue = (niters - peel_iters_prologue) % assumed_vf;
      /* If we need to peel for gaps, but no peeling is required, we have to
	 peel VF iterations.  */
      if (LOOP_VINFO_PEELING_FOR_GAPS (loop_vinfo) && !*peel_iters_epilogue)
	*peel_iters_epilogue = assumed_vf;
    }

  stmt_info_for_cost *si;
  int j;
  if (peel_iters_prologue)
    FOR_EACH_VEC_ELT (*scalar_cost_vec, j, si)
      retval += record_stmt_cost (prologue_cost_vec,
				  si->count * peel_iters_prologue,
				  si->kind, si->stmt_info, si->misalign,
				  vect_prologue);
  if (*peel_iters_epilogue)
    FOR_EACH_VEC_ELT (*scalar_cost_vec, j, si)
      retval += record_stmt_cost (epilogue_cost_vec,
				  si->count * *peel_iters_epilogue,
				  si->kind, si->stmt_info, si->misalign,
				  vect_epilogue);

  return retval;
}

/* Function vect_estimate_min_profitable_iters

   Return the number of iterations required for the vector version of the
   loop to be profitable relative to the cost of the scalar version of the
   loop.

   *RET_MIN_PROFITABLE_NITERS is a cost model profitability threshold
   of iterations for vectorization.  -1 value means loop vectorization
   is not profitable.  This returned value may be used for dynamic
   profitability check.

   *RET_MIN_PROFITABLE_ESTIMATE is a profitability threshold to be used
   for static check against estimated number of iterations.  */

static void
vect_estimate_min_profitable_iters (loop_vec_info loop_vinfo,
				    int *ret_min_profitable_niters,
				    int *ret_min_profitable_estimate)
{
  int min_profitable_iters;
  int min_profitable_estimate;
  int peel_iters_prologue;
  int peel_iters_epilogue;
  unsigned vec_inside_cost = 0;
  int vec_outside_cost = 0;
  unsigned vec_prologue_cost = 0;
  unsigned vec_epilogue_cost = 0;
  int scalar_single_iter_cost = 0;
  int scalar_outside_cost = 0;
  int assumed_vf = vect_vf_for_cost (loop_vinfo);
  int npeel = LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo);
  void *target_cost_data = LOOP_VINFO_TARGET_COST_DATA (loop_vinfo);

  /* Cost model disabled.  */
  if (unlimited_cost_model (LOOP_VINFO_LOOP (loop_vinfo)))
    {
      if (dump_enabled_p ())
	dump_printf_loc (MSG_NOTE, vect_location, "cost model disabled.\n");
      *ret_min_profitable_niters = 0;
      *ret_min_profitable_estimate = 0;
      return;
    }

  /* Requires loop versioning tests to handle misalignment.  */
  if (LOOP_REQUIRES_VERSIONING_FOR_ALIGNMENT (loop_vinfo))
    {
      /*  FIXME: Make cost depend on complexity of individual check.  */
      unsigned len = LOOP_VINFO_MAY_MISALIGN_STMTS (loop_vinfo).length ();
      (void) add_stmt_cost (loop_vinfo, target_cost_data, len, vector_stmt,
			    NULL, NULL_TREE, 0, vect_prologue);
      if (dump_enabled_p ())
	dump_printf (MSG_NOTE,
		     "cost model: Adding cost of checks for loop "
		     "versioning to treat misalignment.\n");
    }

  /* Requires loop versioning with alias checks.  */
  if (LOOP_REQUIRES_VERSIONING_FOR_ALIAS (loop_vinfo))
    {
      /*  FIXME: Make cost depend on complexity of individual check.  */
      unsigned len = LOOP_VINFO_COMP_ALIAS_DDRS (loop_vinfo).length ();
      (void) add_stmt_cost (loop_vinfo, target_cost_data, len, vector_stmt,
			    NULL, NULL_TREE, 0, vect_prologue);
      len = LOOP_VINFO_CHECK_UNEQUAL_ADDRS (loop_vinfo).length ();
      if (len)
	/* Count LEN - 1 ANDs and LEN comparisons.  */
	(void) add_stmt_cost (loop_vinfo, target_cost_data, len * 2 - 1,
			      scalar_stmt, NULL, NULL_TREE, 0, vect_prologue);
      len = LOOP_VINFO_LOWER_BOUNDS (loop_vinfo).length ();
      if (len)
	{
	  /* Count LEN - 1 ANDs and LEN comparisons.  */
	  unsigned int nstmts = len * 2 - 1;
	  /* +1 for each bias that needs adding.  */
	  for (unsigned int i = 0; i < len; ++i)
	    if (!LOOP_VINFO_LOWER_BOUNDS (loop_vinfo)[i].unsigned_p)
	      nstmts += 1;
	  (void) add_stmt_cost (loop_vinfo, target_cost_data, nstmts,
				scalar_stmt, NULL, NULL_TREE, 0, vect_prologue);
	}
      if (dump_enabled_p ())
	dump_printf (MSG_NOTE,
		     "cost model: Adding cost of checks for loop "
		     "versioning aliasing.\n");
    }

  /* Requires loop versioning with niter checks.  */
  if (LOOP_REQUIRES_VERSIONING_FOR_NITERS (loop_vinfo))
    {
      /*  FIXME: Make cost depend on complexity of individual check.  */
      (void) add_stmt_cost (loop_vinfo, target_cost_data, 1, vector_stmt,
			    NULL, NULL_TREE, 0, vect_prologue);
      if (dump_enabled_p ())
	dump_printf (MSG_NOTE,
		     "cost model: Adding cost of checks for loop "
		     "versioning niters.\n");
    }

  if (LOOP_REQUIRES_VERSIONING (loop_vinfo))
    (void) add_stmt_cost (loop_vinfo, target_cost_data, 1, cond_branch_taken,
			  NULL, NULL_TREE, 0, vect_prologue);

  /* Count statements in scalar loop.  Using this as scalar cost for a single
     iteration for now.

     TODO: Add outer loop support.

     TODO: Consider assigning different costs to different scalar
     statements.  */

  scalar_single_iter_cost
    = LOOP_VINFO_SINGLE_SCALAR_ITERATION_COST (loop_vinfo);

  /* Add additional cost for the peeled instructions in prologue and epilogue
     loop.  (For fully-masked loops there will be no peeling.)

     FORNOW: If we don't know the value of peel_iters for prologue or epilogue
     at compile-time - we assume it's vf/2 (the worst would be vf-1).

     TODO: Build an expression that represents peel_iters for prologue and
     epilogue to be used in a run-time test.  */

  if (LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
    {
      peel_iters_prologue = 0;
      peel_iters_epilogue = 0;

      if (LOOP_VINFO_PEELING_FOR_GAPS (loop_vinfo))
	{
	  /* We need to peel exactly one iteration.  */
	  peel_iters_epilogue += 1;
	  stmt_info_for_cost *si;
	  int j;
	  FOR_EACH_VEC_ELT (LOOP_VINFO_SCALAR_ITERATION_COST (loop_vinfo),
			    j, si)
	    (void) add_stmt_cost (loop_vinfo, target_cost_data, si->count,
				  si->kind, si->stmt_info, si->vectype,
				  si->misalign, vect_epilogue);
	}

      /* Calculate how many masks we need to generate.  */
      unsigned int num_masks = 0;
      rgroup_masks *rgm;
      unsigned int num_vectors_m1;
      FOR_EACH_VEC_ELT (LOOP_VINFO_MASKS (loop_vinfo), num_vectors_m1, rgm)
	if (rgm->mask_type)
	  num_masks += num_vectors_m1 + 1;
      gcc_assert (num_masks > 0);

      /* In the worst case, we need to generate each mask in the prologue
	 and in the loop body.  One of the loop body mask instructions
	 replaces the comparison in the scalar loop, and since we don't
	 count the scalar comparison against the scalar body, we shouldn't
	 count that vector instruction against the vector body either.

	 Sometimes we can use unpacks instead of generating prologue
	 masks and sometimes the prologue mask will fold to a constant,
	 so the actual prologue cost might be smaller.  However, it's
	 simpler and safer to use the worst-case cost; if this ends up
	 being the tie-breaker between vectorizing or not, then it's
	 probably better not to vectorize.  */
      (void) add_stmt_cost (loop_vinfo,
			    target_cost_data, num_masks, vector_stmt,
			    NULL, NULL_TREE, 0, vect_prologue);
      (void) add_stmt_cost (loop_vinfo,
			    target_cost_data, num_masks - 1, vector_stmt,
			    NULL, NULL_TREE, 0, vect_body);
    }
  else if (npeel < 0)
    {
      peel_iters_prologue = assumed_vf / 2;
      if (dump_enabled_p ())
	dump_printf (MSG_NOTE, "cost model: "
		     "prologue peel iters set to vf/2.\n");

      /* If peeling for alignment is unknown, loop bound of main loop becomes
         unknown.  */
      peel_iters_epilogue = assumed_vf / 2;
      if (dump_enabled_p ())
	dump_printf (MSG_NOTE, "cost model: "
		     "epilogue peel iters set to vf/2 because "
		     "peeling for alignment is unknown.\n");

      /* If peeled iterations are unknown, count a taken branch and a not taken
         branch per peeled loop. Even if scalar loop iterations are known,
         vector iterations are not known since peeled prologue iterations are
         not known. Hence guards remain the same.  */
      (void) add_stmt_cost (loop_vinfo, target_cost_data, 1, cond_branch_taken,
			    NULL, NULL_TREE, 0, vect_prologue);
      (void) add_stmt_cost (loop_vinfo,
			    target_cost_data, 1, cond_branch_not_taken,
			    NULL, NULL_TREE, 0, vect_prologue);
      (void) add_stmt_cost (loop_vinfo, target_cost_data, 1, cond_branch_taken,
			    NULL, NULL_TREE, 0, vect_epilogue);
      (void) add_stmt_cost (loop_vinfo,
			    target_cost_data, 1, cond_branch_not_taken,
			    NULL, NULL_TREE, 0, vect_epilogue);
      stmt_info_for_cost *si;
      int j;
      FOR_EACH_VEC_ELT (LOOP_VINFO_SCALAR_ITERATION_COST (loop_vinfo), j, si)
	{
	  (void) add_stmt_cost (loop_vinfo, target_cost_data,
				si->count * peel_iters_prologue,
				si->kind, si->stmt_info, si->vectype,
				si->misalign,
				vect_prologue);
	  (void) add_stmt_cost (loop_vinfo, target_cost_data,
				si->count * peel_iters_epilogue,
				si->kind, si->stmt_info, si->vectype,
				si->misalign,
				vect_epilogue);
	}
    }
  else
    {
      stmt_vector_for_cost prologue_cost_vec, epilogue_cost_vec;
      stmt_info_for_cost *si;
      int j;
      void *data = LOOP_VINFO_TARGET_COST_DATA (loop_vinfo);

      prologue_cost_vec.create (2);
      epilogue_cost_vec.create (2);
      peel_iters_prologue = npeel;

      (void) vect_get_known_peeling_cost (loop_vinfo, peel_iters_prologue,
					  &peel_iters_epilogue,
					  &LOOP_VINFO_SCALAR_ITERATION_COST
					    (loop_vinfo),
					  &prologue_cost_vec,
					  &epilogue_cost_vec);

      FOR_EACH_VEC_ELT (prologue_cost_vec, j, si)
	(void) add_stmt_cost (loop_vinfo,
			      data, si->count, si->kind, si->stmt_info,
			      si->vectype, si->misalign, vect_prologue);

      FOR_EACH_VEC_ELT (epilogue_cost_vec, j, si)
	(void) add_stmt_cost (loop_vinfo,
			      data, si->count, si->kind, si->stmt_info,
			      si->vectype, si->misalign, vect_epilogue);

      prologue_cost_vec.release ();
      epilogue_cost_vec.release ();
    }

  /* FORNOW: The scalar outside cost is incremented in one of the
     following ways:

     1. The vectorizer checks for alignment and aliasing and generates
     a condition that allows dynamic vectorization.  A cost model
     check is ANDED with the versioning condition.  Hence scalar code
     path now has the added cost of the versioning check.

       if (cost > th & versioning_check)
         jmp to vector code

     Hence run-time scalar is incremented by not-taken branch cost.

     2. The vectorizer then checks if a prologue is required.  If the
     cost model check was not done before during versioning, it has to
     be done before the prologue check.

       if (cost <= th)
         prologue = scalar_iters
       if (prologue == 0)
         jmp to vector code
       else
         execute prologue
       if (prologue == num_iters)
	 go to exit

     Hence the run-time scalar cost is incremented by a taken branch,
     plus a not-taken branch, plus a taken branch cost.

     3. The vectorizer then checks if an epilogue is required.  If the
     cost model check was not done before during prologue check, it
     has to be done with the epilogue check.

       if (prologue == 0)
         jmp to vector code
       else
         execute prologue
       if (prologue == num_iters)
	 go to exit
       vector code:
         if ((cost <= th) | (scalar_iters-prologue-epilogue == 0))
           jmp to epilogue

     Hence the run-time scalar cost should be incremented by 2 taken
     branches.

     TODO: The back end may reorder the BBS's differently and reverse
     conditions/branch directions.  Change the estimates below to
     something more reasonable.  */

  /* If the number of iterations is known and we do not do versioning, we can
     decide whether to vectorize at compile time.  Hence the scalar version
     do not carry cost model guard costs.  */
  if (!LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo)
      || LOOP_REQUIRES_VERSIONING (loop_vinfo))
    {
      /* Cost model check occurs at versioning.  */
      if (LOOP_REQUIRES_VERSIONING (loop_vinfo))
	scalar_outside_cost += vect_get_stmt_cost (cond_branch_not_taken);
      else
	{
	  /* Cost model check occurs at prologue generation.  */
	  if (LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo) < 0)
	    scalar_outside_cost += 2 * vect_get_stmt_cost (cond_branch_taken)
	      + vect_get_stmt_cost (cond_branch_not_taken); 
	  /* Cost model check occurs at epilogue generation.  */
	  else
	    scalar_outside_cost += 2 * vect_get_stmt_cost (cond_branch_taken); 
	}
    }

  /* Complete the target-specific cost calculations.  */
  finish_cost (LOOP_VINFO_TARGET_COST_DATA (loop_vinfo), &vec_prologue_cost,
	       &vec_inside_cost, &vec_epilogue_cost);

  vec_outside_cost = (int)(vec_prologue_cost + vec_epilogue_cost);

  /* Stash the costs so that we can compare two loop_vec_infos.  */
  loop_vinfo->vec_inside_cost = vec_inside_cost;
  loop_vinfo->vec_outside_cost = vec_outside_cost;

  if (dump_enabled_p ())
    {
      dump_printf_loc (MSG_NOTE, vect_location, "Cost model analysis: \n");
      dump_printf (MSG_NOTE, "  Vector inside of loop cost: %d\n",
                   vec_inside_cost);
      dump_printf (MSG_NOTE, "  Vector prologue cost: %d\n",
                   vec_prologue_cost);
      dump_printf (MSG_NOTE, "  Vector epilogue cost: %d\n",
                   vec_epilogue_cost);
      dump_printf (MSG_NOTE, "  Scalar iteration cost: %d\n",
                   scalar_single_iter_cost);
      dump_printf (MSG_NOTE, "  Scalar outside cost: %d\n",
                   scalar_outside_cost);
      dump_printf (MSG_NOTE, "  Vector outside cost: %d\n",
                   vec_outside_cost);
      dump_printf (MSG_NOTE, "  prologue iterations: %d\n",
                   peel_iters_prologue);
      dump_printf (MSG_NOTE, "  epilogue iterations: %d\n",
                   peel_iters_epilogue);
    }

  /* Calculate number of iterations required to make the vector version
     profitable, relative to the loop bodies only.  The following condition
     must hold true:
     SIC * niters + SOC > VIC * ((niters - NPEEL) / VF) + VOC
     where
     SIC = scalar iteration cost, VIC = vector iteration cost,
     VOC = vector outside cost, VF = vectorization factor,
     NPEEL = prologue iterations + epilogue iterations,
     SOC = scalar outside cost for run time cost model check.  */

  int saving_per_viter = (scalar_single_iter_cost * assumed_vf
			  - vec_inside_cost);
  if (saving_per_viter <= 0)
    {
      if (LOOP_VINFO_LOOP (loop_vinfo)->force_vectorize)
	warning_at (vect_location.get_location_t (), OPT_Wopenmp_simd,
		    "vectorization did not happen for a simd loop");

      if (dump_enabled_p ())
        dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
			 "cost model: the vector iteration cost = %d "
			 "divided by the scalar iteration cost = %d "
			 "is greater or equal to the vectorization factor = %d"
                         ".\n",
			 vec_inside_cost, scalar_single_iter_cost, assumed_vf);
      *ret_min_profitable_niters = -1;
      *ret_min_profitable_estimate = -1;
      return;
    }

  /* ??? The "if" arm is written to handle all cases; see below for what
     we would do for !LOOP_VINFO_FULLY_MASKED_P.  */
  if (LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
    {
      /* Rewriting the condition above in terms of the number of
	 vector iterations (vniters) rather than the number of
	 scalar iterations (niters) gives:

	 SIC * (vniters * VF + NPEEL) + SOC > VIC * vniters + VOC

	 <==> vniters * (SIC * VF - VIC) > VOC - SIC * NPEEL - SOC

	 For integer N, X and Y when X > 0:

	 N * X > Y <==> N >= (Y /[floor] X) + 1.  */
      int outside_overhead = (vec_outside_cost
			      - scalar_single_iter_cost * peel_iters_prologue
			      - scalar_single_iter_cost * peel_iters_epilogue
			      - scalar_outside_cost);
      /* We're only interested in cases that require at least one
	 vector iteration.  */
      int min_vec_niters = 1;
      if (outside_overhead > 0)
	min_vec_niters = outside_overhead / saving_per_viter + 1;

      if (dump_enabled_p ())
	dump_printf (MSG_NOTE, "  Minimum number of vector iterations: %d\n",
		     min_vec_niters);

      if (LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
	{
	  /* Now that we know the minimum number of vector iterations,
	     find the minimum niters for which the scalar cost is larger:

	     SIC * niters > VIC * vniters + VOC - SOC

	     We know that the minimum niters is no more than
	     vniters * VF + NPEEL, but it might be (and often is) less
	     than that if a partial vector iteration is cheaper than the
	     equivalent scalar code.  */
	  int threshold = (vec_inside_cost * min_vec_niters
			   + vec_outside_cost
			   - scalar_outside_cost);
	  if (threshold <= 0)
	    min_profitable_iters = 1;
	  else
	    min_profitable_iters = threshold / scalar_single_iter_cost + 1;
	}
      else
	/* Convert the number of vector iterations into a number of
	   scalar iterations.  */
	min_profitable_iters = (min_vec_niters * assumed_vf
				+ peel_iters_prologue
				+ peel_iters_epilogue);
    }
  else
    {
      min_profitable_iters = ((vec_outside_cost - scalar_outside_cost)
			      * assumed_vf
			      - vec_inside_cost * peel_iters_prologue
			      - vec_inside_cost * peel_iters_epilogue);
      if (min_profitable_iters <= 0)
        min_profitable_iters = 0;
      else
	{
	  min_profitable_iters /= saving_per_viter;

	  if ((scalar_single_iter_cost * assumed_vf * min_profitable_iters)
	      <= (((int) vec_inside_cost * min_profitable_iters)
		  + (((int) vec_outside_cost - scalar_outside_cost)
		     * assumed_vf)))
	    min_profitable_iters++;
	}
    }

  if (dump_enabled_p ())
    dump_printf (MSG_NOTE,
		 "  Calculated minimum iters for profitability: %d\n",
		 min_profitable_iters);

  if (!LOOP_VINFO_FULLY_MASKED_P (loop_vinfo)
      && min_profitable_iters < (assumed_vf + peel_iters_prologue))
    /* We want the vectorized loop to execute at least once.  */
    min_profitable_iters = assumed_vf + peel_iters_prologue;

  if (dump_enabled_p ())
    dump_printf_loc (MSG_NOTE, vect_location,
                     "  Runtime profitability threshold = %d\n",
                     min_profitable_iters);

  *ret_min_profitable_niters = min_profitable_iters;

  /* Calculate number of iterations required to make the vector version
     profitable, relative to the loop bodies only.

     Non-vectorized variant is SIC * niters and it must win over vector
     variant on the expected loop trip count.  The following condition must hold true:
     SIC * niters > VIC * ((niters - NPEEL) / VF) + VOC + SOC  */

  if (vec_outside_cost <= 0)
    min_profitable_estimate = 0;
  else if (LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
    {
      /* This is a repeat of the code above, but with + SOC rather
	 than - SOC.  */
      int outside_overhead = (vec_outside_cost
			      - scalar_single_iter_cost * peel_iters_prologue
			      - scalar_single_iter_cost * peel_iters_epilogue
			      + scalar_outside_cost);
      int min_vec_niters = 1;
      if (outside_overhead > 0)
	min_vec_niters = outside_overhead / saving_per_viter + 1;

      if (LOOP_VINFO_FULLY_MASKED_P (loop_vinfo))
	{
	  int threshold = (vec_inside_cost * min_vec_niters
			   + vec_outside_cost
			   + scalar_outside_cost);
	  min_profitable_estimate = threshold / scalar_single_iter_cost + 1;
	}
      else
	min_profitable_estimate = (min_vec_niters * assumed_vf
				   + peel_iters_prologue
				   + peel_iters_epilogue);
    }
  else
    {
      min_profitable_estimate = ((vec_outside_cost + scalar_outside_cost)
				 * assumed_vf
				 - vec_inside_cost * peel_iters_prologue
				 - vec_inside_cost * peel_iters_epilogue)
				 / ((scalar_single_iter_cost * assumed_vf)
				   - vec_inside_cost);
    }
  min_profitable_estimate = MAX (min_profitable_estimate, min_profitable_iters);
  if (dump_enabled_p ())
    dump_printf_loc (MSG_NOTE, vect_location,
		     "  Static estimate profitability threshold = %d\n",
		     min_profitable_estimate);

  *ret_min_profitable_estimate = min_profitable_estimate;
}

/* Writes into SEL a mask for a vec_perm, equivalent to a vec_shr by OFFSET
   vector elements (not bits) for a vector with NELT elements.  */
static void
calc_vec_perm_mask_for_shift (unsigned int offset, unsigned int nelt,
			      vec_perm_builder *sel)
{
  /* The encoding is a single stepped pattern.  Any wrap-around is handled
     by vec_perm_indices.  */
  sel->new_vector (nelt, 1, 3);
  for (unsigned int i = 0; i < 3; i++)
    sel->quick_push (i + offset);
}

/* Checks whether the target supports whole-vector shifts for vectors of mode
   MODE.  This is the case if _either_ the platform handles vec_shr_optab, _or_
   it supports vec_perm_const with masks for all necessary shift amounts.  */
static bool
have_whole_vector_shift (machine_mode mode)
{
  if (optab_handler (vec_shr_optab, mode) != CODE_FOR_nothing)
    return true;

  /* Variable-length vectors should be handled via the optab.  */
  unsigned int nelt;
  if (!GET_MODE_NUNITS (mode).is_constant (&nelt))
    return false;

  vec_perm_builder sel;
  vec_perm_indices indices;
  for (unsigned int i = nelt / 2; i >= 1; i /= 2)
    {
      calc_vec_perm_mask_for_shift (i, nelt, &sel);
      indices.new_vector (sel, 2, nelt);
      if (!can_vec_perm_const_p (mode, indices, false))
	return false;
    }
  return true;
}

/* TODO: Close dependency between vect_model_*_cost and vectorizable_*
   functions. Design better to avoid maintenance issues.  */

/* Function vect_model_reduction_cost.

   Models cost for a reduction operation, including the vector ops
   generated within the strip-mine loop, the initial definition before
   the loop, and the epilogue code that must be generated.  */

static void
vect_model_reduction_cost (loop_vec_info loop_vinfo,
			   stmt_vec_info stmt_info, internal_fn reduc_fn,
			   vect_reduction_type reduction_type,
			   int ncopies, stmt_vector_for_cost *cost_vec)
{
  int prologue_cost = 0, epilogue_cost = 0, inside_cost;
  enum tree_code code;
  optab optab;
  tree vectype;
  machine_mode mode;
  class loop *loop = NULL;

  if (loop_vinfo)
    loop = LOOP_VINFO_LOOP (loop_vinfo);

  /* Condition reductions generate two reductions in the loop.  */
  if (reduction_type == COND_REDUCTION)
    ncopies *= 2;

  vectype = STMT_VINFO_VECTYPE (stmt_info);
  mode = TYPE_MODE (vectype);
  stmt_vec_info orig_stmt_info = vect_orig_stmt (stmt_info);

  code = gimple_assign_rhs_code (orig_stmt_info->stmt);

  if (reduction_type == EXTRACT_LAST_REDUCTION)
    /* No extra instructions are needed in the prologue.  The loop body
       operations are costed in vectorizable_condition.  */
    inside_cost = 0;
  else if (reduction_type == FOLD_LEFT_REDUCTION)
    {
      /* No extra instructions needed in the prologue.  */
      prologue_cost = 0;

      if (reduc_fn != IFN_LAST)
	/* Count one reduction-like operation per vector.  */
	inside_cost = record_stmt_cost (cost_vec, ncopies, vec_to_scalar,
					stmt_info, 0, vect_body);
      else
	{
	  /* Use NELEMENTS extracts and NELEMENTS scalar ops.  */
	  unsigned int nelements = ncopies * vect_nunits_for_cost (vectype);
	  inside_cost = record_stmt_cost (cost_vec, nelements,
					  vec_to_scalar, stmt_info, 0,
					  vect_body);
	  inside_cost += record_stmt_cost (cost_vec, nelements,
					   scalar_stmt, stmt_info, 0,
					   vect_body);
	}
    }
  else
    {
      /* Add in cost for initial definition.
	 For cond reduction we have four vectors: initial index, step,
	 initial result of the data reduction, initial value of the index
	 reduction.  */
      int prologue_stmts = reduction_type == COND_REDUCTION ? 4 : 1;
      prologue_cost += record_stmt_cost (cost_vec, prologue_stmts,
					 scalar_to_vec, stmt_info, 0,
					 vect_prologue);

      /* Cost of reduction op inside loop.  */
      inside_cost = record_stmt_cost (cost_vec, ncopies, vector_stmt,
				      stmt_info, 0, vect_body);
    }

  /* Determine cost of epilogue code.

     We have a reduction operator that will reduce the vector in one statement.
     Also requires scalar extract.  */

  if (!loop || !nested_in_vect_loop_p (loop, orig_stmt_info))
    {
      if (reduc_fn != IFN_LAST)
	{
	  if (reduction_type == COND_REDUCTION)
	    {
	      /* An EQ stmt and an COND_EXPR stmt.  */
	      epilogue_cost += record_stmt_cost (cost_vec, 2,
						 vector_stmt, stmt_info, 0,
						 vect_epilogue);
	      /* Reduction of the max index and a reduction of the found
		 values.  */
	      epilogue_cost += record_stmt_cost (cost_vec, 2,
						 vec_to_scalar, stmt_info, 0,
						 vect_epilogue);
	      /* A broadcast of the max value.  */
	      epilogue_cost += record_stmt_cost (cost_vec, 1,
						 scalar_to_vec, stmt_info, 0,
						 vect_epilogue);
	    }
	  else
	    {
	      epilogue_cost += record_stmt_cost (cost_vec, 1, vector_stmt,
						 stmt_info, 0, vect_epilogue);
	      epilogue_cost += record_stmt_cost (cost_vec, 1,
						 vec_to_scalar, stmt_info, 0,
						 vect_epilogue);
	    }
	}
      else if (reduction_type == COND_REDUCTION)
	{
	  unsigned estimated_nunits = vect_nunits_for_cost (vectype);
	  /* Extraction of scalar elements.  */
	  epilogue_cost += record_stmt_cost (cost_vec,
					     2 * estimated_nunits,
					     vec_to_scalar, stmt_info, 0,
					     vect_epilogue);
	  /* Scalar max reductions via COND_EXPR / MAX_EXPR.  */
	  epilogue_cost += record_stmt_cost (cost_vec,
					     2 * estimated_nunits - 3,
					     scalar_stmt, stmt_info, 0,
					     vect_epilogue);
	}
      else if (reduction_type == EXTRACT_LAST_REDUCTION
	       || reduction_type == FOLD_LEFT_REDUCTION)
	/* No extra instructions need in the epilogue.  */
	;
      else
	{
	  int vec_size_in_bits = tree_to_uhwi (TYPE_SIZE (vectype));
	  tree bitsize =
	    TYPE_SIZE (TREE_TYPE (gimple_assign_lhs (orig_stmt_info->stmt)));
	  int element_bitsize = tree_to_uhwi (bitsize);
	  int nelements = vec_size_in_bits / element_bitsize;

	  if (code == COND_EXPR)
	    code = MAX_EXPR;

	  optab = optab_for_tree_code (code, vectype, optab_default);

	  /* We have a whole vector shift available.  */
	  if (optab != unknown_optab
	      && VECTOR_MODE_P (mode)
	      && optab_handler (optab, mode) != CODE_FOR_nothing
	      && have_whole_vector_shift (mode))
	    {
	      /* Final reduction via vector shifts and the reduction operator.
		 Also requires scalar extract.  */
	      epilogue_cost += record_stmt_cost (cost_vec,
						 exact_log2 (nelements) * 2,
						 vector_stmt, stmt_info, 0,
						 vect_epilogue);
	      epilogue_cost += record_stmt_cost (cost_vec, 1,
						 vec_to_scalar, stmt_info, 0,
						 vect_epilogue);
	    }	  
	  else
	    /* Use extracts and reduction op for final reduction.  For N
	       elements, we have N extracts and N-1 reduction ops.  */
	    epilogue_cost += record_stmt_cost (cost_vec, 
					       nelements + nelements - 1,
					       vector_stmt, stmt_info, 0,
					       vect_epilogue);
	}
    }

  if (dump_enabled_p ())
    dump_printf (MSG_NOTE, 
                 "vect_model_reduction_cost: inside_cost = %d, "
                 "prologue_cost = %d, epilogue_cost = %d .\n", inside_cost,
                 prologue_cost, epilogue_cost);
}


/* Function vect_model_induction_cost.

   Models cost for induction operations.  */

static void
vect_model_induction_cost (stmt_vec_info stmt_info, int ncopies,
			   stmt_vector_for_cost *cost_vec)
{
  unsigned inside_cost, prologue_cost;

  if (PURE_SLP_STMT (stmt_info))
    return;

  /* loop cost for vec_loop.  */
  inside_cost = record_stmt_cost (cost_vec, ncopies, vector_stmt,
				  stmt_info, 0, vect_body);

  /* prologue cost for vec_init and vec_step.  */
  prologue_cost = record_stmt_cost (cost_vec, 2, scalar_to_vec,
				    stmt_info, 0, vect_prologue);

  if (dump_enabled_p ())
    dump_printf_loc (MSG_NOTE, vect_location,
                     "vect_model_induction_cost: inside_cost = %d, "
                     "prologue_cost = %d .\n", inside_cost, prologue_cost);
}



/* Function get_initial_def_for_reduction

   Input:
   STMT_VINFO - a stmt that performs a reduction operation in the loop.
   INIT_VAL - the initial value of the reduction variable

   Output:
   ADJUSTMENT_DEF - a tree that holds a value to be added to the final result
        of the reduction (used for adjusting the epilog - see below).
   Return a vector variable, initialized according to the operation that
	STMT_VINFO performs. This vector will be used as the initial value
	of the vector of partial results.

   Option1 (adjust in epilog): Initialize the vector as follows:
     add/bit or/xor:    [0,0,...,0,0]
     mult/bit and:      [1,1,...,1,1]
     min/max/cond_expr: [init_val,init_val,..,init_val,init_val]
   and when necessary (e.g. add/mult case) let the caller know
   that it needs to adjust the result by init_val.

   Option2: Initialize the vector as follows:
     add/bit or/xor:    [init_val,0,0,...,0]
     mult/bit and:      [init_val,1,1,...,1]
     min/max/cond_expr: [init_val,init_val,...,init_val]
   and no adjustments are needed.

   For example, for the following code:

   s = init_val;
   for (i=0;i<n;i++)
     s = s + a[i];

   STMT_VINFO is 's = s + a[i]', and the reduction variable is 's'.
   For a vector of 4 units, we want to return either [0,0,0,init_val],
   or [0,0,0,0] and let the caller know that it needs to adjust
   the result at the end by 'init_val'.

   FORNOW, we are using the 'adjust in epilog' scheme, because this way the
   initialization vector is simpler (same element in all entries), if
   ADJUSTMENT_DEF is not NULL, and Option2 otherwise.

   A cost model should help decide between these two schemes.  */

static tree
get_initial_def_for_reduction (loop_vec_info loop_vinfo,
			       stmt_vec_info stmt_vinfo,
			       enum tree_code code, tree init_val,
                               tree *adjustment_def)
{
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo);
  tree scalar_type = TREE_TYPE (init_val);
  tree vectype = get_vectype_for_scalar_type (loop_vinfo, scalar_type);
  tree def_for_init;
  tree init_def;
  REAL_VALUE_TYPE real_init_val = dconst0;
  int int_init_val = 0;
  gimple_seq stmts = NULL;

  gcc_assert (vectype);

  gcc_assert (POINTER_TYPE_P (scalar_type) || INTEGRAL_TYPE_P (scalar_type)
	      || SCALAR_FLOAT_TYPE_P (scalar_type));

  gcc_assert (nested_in_vect_loop_p (loop, stmt_vinfo)
	      || loop == (gimple_bb (stmt_vinfo->stmt))->loop_father);

  /* ADJUSTMENT_DEF is NULL when called from
     vect_create_epilog_for_reduction to vectorize double reduction.  */
  if (adjustment_def)
    *adjustment_def = NULL;

  switch (code)
    {
    case WIDEN_SUM_EXPR:
    case DOT_PROD_EXPR:
    case SAD_EXPR:
    case PLUS_EXPR:
    case MINUS_EXPR:
    case BIT_IOR_EXPR:
    case BIT_XOR_EXPR:
    case MULT_EXPR:
    case BIT_AND_EXPR:
      {
        if (code == MULT_EXPR)
          {
            real_init_val = dconst1;
            int_init_val = 1;
          }

        if (code == BIT_AND_EXPR)
          int_init_val = -1;

        if (SCALAR_FLOAT_TYPE_P (scalar_type))
          def_for_init = build_real (scalar_type, real_init_val);
        else
          def_for_init = build_int_cst (scalar_type, int_init_val);

	if (adjustment_def || operand_equal_p (def_for_init, init_val, 0))
	  {
	    /* Option1: the first element is '0' or '1' as well.  */
	    if (!operand_equal_p (def_for_init, init_val, 0))
	      *adjustment_def = init_val;
	    init_def = gimple_build_vector_from_val (&stmts, vectype,
						     def_for_init);
	  }
	else if (!TYPE_VECTOR_SUBPARTS (vectype).is_constant ())
	  {
	    /* Option2 (variable length): the first element is INIT_VAL.  */
	    init_def = gimple_build_vector_from_val (&stmts, vectype,
						     def_for_init);
	    init_def = gimple_build (&stmts, CFN_VEC_SHL_INSERT,
				     vectype, init_def, init_val);
	  }
	else
	  {
	    /* Option2: the first element is INIT_VAL.  */
	    tree_vector_builder elts (vectype, 1, 2);
	    elts.quick_push (init_val);
	    elts.quick_push (def_for_init);
	    init_def = gimple_build_vector (&stmts, &elts);
	  }
      }
      break;

    case MIN_EXPR:
    case MAX_EXPR:
    case COND_EXPR:
      {
	init_val = gimple_convert (&stmts, TREE_TYPE (vectype), init_val);
	init_def = gimple_build_vector_from_val (&stmts, vectype, init_val);
      }
      break;

    default:
      gcc_unreachable ();
    }

  if (stmts)
    gsi_insert_seq_on_edge_immediate (loop_preheader_edge (loop), stmts);
  return init_def;
}

/* Get at the initial defs for the reduction PHIs in SLP_NODE.
   NUMBER_OF_VECTORS is the number of vector defs to create.
   If NEUTRAL_OP is nonnull, introducing extra elements of that
   value will not change the result.  */

static void
get_initial_defs_for_reduction (vec_info *vinfo,
				slp_tree slp_node,
				vec<tree> *vec_oprnds,
				unsigned int number_of_vectors,
				bool reduc_chain, tree neutral_op)
{
  vec<stmt_vec_info> stmts = SLP_TREE_SCALAR_STMTS (slp_node);
  stmt_vec_info stmt_vinfo = stmts[0];
  unsigned HOST_WIDE_INT nunits;
  unsigned j, number_of_places_left_in_vector;
  tree vector_type;
  unsigned int group_size = stmts.length ();
  unsigned int i;
  class loop *loop;

  vector_type = STMT_VINFO_VECTYPE (stmt_vinfo);

  gcc_assert (STMT_VINFO_DEF_TYPE (stmt_vinfo) == vect_reduction_def);

  loop = (gimple_bb (stmt_vinfo->stmt))->loop_father;
  gcc_assert (loop);
  edge pe = loop_preheader_edge (loop);

  gcc_assert (!reduc_chain || neutral_op);

  /* NUMBER_OF_COPIES is the number of times we need to use the same values in
     created vectors. It is greater than 1 if unrolling is performed.

     For example, we have two scalar operands, s1 and s2 (e.g., group of
     strided accesses of size two), while NUNITS is four (i.e., four scalars
     of this type can be packed in a vector).  The output vector will contain
     two copies of each scalar operand: {s1, s2, s1, s2}.  (NUMBER_OF_COPIES
     will be 2).

     If REDUC_GROUP_SIZE > NUNITS, the scalars will be split into several
     vectors containing the operands.

     For example, NUNITS is four as before, and the group size is 8
     (s1, s2, ..., s8).  We will create two vectors {s1, s2, s3, s4} and
     {s5, s6, s7, s8}.  */

  if (!TYPE_VECTOR_SUBPARTS (vector_type).is_constant (&nunits))
    nunits = group_size;

  number_of_places_left_in_vector = nunits;
  bool constant_p = true;
  tree_vector_builder elts (vector_type, nunits, 1);
  elts.quick_grow (nunits);
  gimple_seq ctor_seq = NULL;
  for (j = 0; j < nunits * number_of_vectors; ++j)
    {
      tree op;
      i = j % group_size;
      stmt_vinfo = stmts[i];

      /* Get the def before the loop.  In reduction chain we have only
	 one initial value.  Else we have as many as PHIs in the group.  */
      if (reduc_chain)
	op = j != 0 ? neutral_op : PHI_ARG_DEF_FROM_EDGE (stmt_vinfo->stmt, pe);
      else if (((vec_oprnds->length () + 1) * nunits
		- number_of_places_left_in_vector >= group_size)
	       && neutral_op)
	op = neutral_op;
      else
	op = PHI_ARG_DEF_FROM_EDGE (stmt_vinfo->stmt, pe);

      /* Create 'vect_ = {op0,op1,...,opn}'.  */
      number_of_places_left_in_vector--;
      elts[nunits - number_of_places_left_in_vector - 1] = op;
      if (!CONSTANT_CLASS_P (op))
	constant_p = false;

      if (number_of_places_left_in_vector == 0)
	{
	  tree init;
	  if (constant_p && !neutral_op
	      ? multiple_p (TYPE_VECTOR_SUBPARTS (vector_type), nunits)
	      : known_eq (TYPE_VECTOR_SUBPARTS (vector_type), nunits))
	    /* Build the vector directly from ELTS.  */
	    init = gimple_build_vector (&ctor_seq, &elts);
	  else if (neutral_op)
	    {
	      /* Build a vector of the neutral value and shift the
		 other elements into place.  */
	      init = gimple_build_vector_from_val (&ctor_seq, vector_type,
						   neutral_op);
	      int k = nunits;
	      while (k > 0 && elts[k - 1] == neutral_op)
		k -= 1;
	      while (k > 0)
		{
		  k -= 1;
		  init = gimple_build (&ctor_seq, CFN_VEC_SHL_INSERT,
				       vector_type, init, elts[k]);
		}
	    }
	  else
	    {
	      /* First time round, duplicate ELTS to fill the
		 required number of vectors.  */
	      duplicate_and_interleave (vinfo, &ctor_seq, vector_type, elts,
					number_of_vectors, *vec_oprnds);
	      break;
	    }
	  vec_oprnds->quick_push (init);

	  number_of_places_left_in_vector = nunits;
	  elts.new_vector (vector_type, nunits, 1);
	  elts.quick_grow (nunits);
	  constant_p = true;
	}
    }
  if (ctor_seq != NULL)
    gsi_insert_seq_on_edge_immediate (pe, ctor_seq);
}

/* For a statement STMT_INFO taking part in a reduction operation return
   the stmt_vec_info the meta information is stored on.  */

stmt_vec_info
info_for_reduction (vec_info *vinfo, stmt_vec_info stmt_info)
{
  stmt_info = vect_orig_stmt (stmt_info);
  gcc_assert (STMT_VINFO_REDUC_DEF (stmt_info));
  if (!is_a <gphi *> (stmt_info->stmt))
    stmt_info = STMT_VINFO_REDUC_DEF (stmt_info);
  gphi *phi = as_a <gphi *> (stmt_info->stmt);
  if (STMT_VINFO_DEF_TYPE (stmt_info) == vect_double_reduction_def)
    {
      if (gimple_phi_num_args (phi) == 1)
	stmt_info = STMT_VINFO_REDUC_DEF (stmt_info);
    }
  else if (STMT_VINFO_DEF_TYPE (stmt_info) == vect_nested_cycle)
    {
      edge pe = loop_preheader_edge (gimple_bb (phi)->loop_father);
      stmt_vec_info info
	  = vinfo->lookup_def (PHI_ARG_DEF_FROM_EDGE (phi, pe));
      if (info && STMT_VINFO_DEF_TYPE (info) == vect_double_reduction_def)
	stmt_info = info;
    }
  return stmt_info;
}

/* Function vect_create_epilog_for_reduction

   Create code at the loop-epilog to finalize the result of a reduction
   computation. 
  
   STMT_INFO is the scalar reduction stmt that is being vectorized.
   SLP_NODE is an SLP node containing a group of reduction statements. The 
     first one in this group is STMT_INFO.
   SLP_NODE_INSTANCE is the SLP node instance containing SLP_NODE
   REDUC_INDEX says which rhs operand of the STMT_INFO is the reduction phi
     (counting from 0)

   This function:
   1. Completes the reduction def-use cycles.
   2. "Reduces" each vector of partial results VECT_DEFS into a single result,
      by calling the function specified by REDUC_FN if available, or by
      other means (whole-vector shifts or a scalar loop).
      The function also creates a new phi node at the loop exit to preserve
      loop-closed form, as illustrated below.

     The flow at the entry to this function:

        loop:
          vec_def = phi <vec_init, null>        # REDUCTION_PHI
          VECT_DEF = vector_stmt                # vectorized form of STMT_INFO
          s_loop = scalar_stmt                  # (scalar) STMT_INFO
        loop_exit:
          s_out0 = phi <s_loop>                 # (scalar) EXIT_PHI
          use <s_out0>
          use <s_out0>

     The above is transformed by this function into:

        loop:
          vec_def = phi <vec_init, VECT_DEF>    # REDUCTION_PHI
          VECT_DEF = vector_stmt                # vectorized form of STMT_INFO
          s_loop = scalar_stmt                  # (scalar) STMT_INFO
        loop_exit:
          s_out0 = phi <s_loop>                 # (scalar) EXIT_PHI
          v_out1 = phi <VECT_DEF>               # NEW_EXIT_PHI
          v_out2 = reduce <v_out1>
          s_out3 = extract_field <v_out2, 0>
          s_out4 = adjust_result <s_out3>
          use <s_out4>
          use <s_out4>
*/

static void
vect_create_epilog_for_reduction (loop_vec_info loop_vinfo,
				  stmt_vec_info stmt_info,
				  slp_tree slp_node,
				  slp_instance slp_node_instance)
{
  stmt_vec_info reduc_info = info_for_reduction (loop_vinfo, stmt_info);
  gcc_assert (reduc_info->is_reduc_info);
  /* For double reductions we need to get at the inner loop reduction
     stmt which has the meta info attached.  Our stmt_info is that of the
     loop-closed PHI of the inner loop which we remember as
     def for the reduction PHI generation.  */
  bool double_reduc = false;
  stmt_vec_info rdef_info = stmt_info;
  if (STMT_VINFO_DEF_TYPE (stmt_info) == vect_double_reduction_def)
    {
      gcc_assert (!slp_node);
      double_reduc = true;
      stmt_info = loop_vinfo->lookup_def (gimple_phi_arg_def
					    (stmt_info->stmt, 0));
      stmt_info = vect_stmt_to_vectorize (stmt_info);
    }
  gphi *reduc_def_stmt
    = as_a <gphi *> (STMT_VINFO_REDUC_DEF (vect_orig_stmt (stmt_info))->stmt);
  enum tree_code code = STMT_VINFO_REDUC_CODE (reduc_info);
  internal_fn reduc_fn = STMT_VINFO_REDUC_FN (reduc_info);
  stmt_vec_info prev_phi_info;
  tree vectype;
  machine_mode mode;
  class loop *loop = LOOP_VINFO_LOOP (loop_vinfo), *outer_loop = NULL;
  basic_block exit_bb;
  tree scalar_dest;
  tree scalar_type;
  gimple *new_phi = NULL, *phi;
  stmt_vec_info phi_info;
  gimple_stmt_iterator exit_gsi;
  tree new_temp = NULL_TREE, new_name, new_scalar_dest;
  gimple *epilog_stmt = NULL;
  gimple *exit_phi;
  tree bitsize;
  tree def;
  tree orig_name, scalar_result;
  imm_use_iterator imm_iter, phi_imm_iter;
  use_operand_p use_p, phi_use_p;
  gimple *use_stmt;
  bool nested_in_vect_loop = false;
  auto_vec<gimple *> new_phis;
  int j, i;
  auto_vec<tree> scalar_results;
  unsigned int group_size = 1, k;
  auto_vec<gimple *> phis;
  bool slp_reduc = false;
  bool direct_slp_reduc;
  tree new_phi_result;
  tree induction_index = NULL_TREE;

  if (slp_node)
    group_size = SLP_TREE_LANES (slp_node);

  if (nested_in_vect_loop_p (loop, stmt_info))
    {
      outer_loop = loop;
      loop = loop->inner;
      nested_in_vect_loop = true;
      gcc_assert (!slp_node);
    }
  gcc_assert (!nested_in_vect_loop || double_reduc);

  vectype = STMT_VINFO_REDUC_VECTYPE (reduc_info);
  gcc_assert (vectype);
  mode = TYPE_MODE (vectype);

  tree initial_def = NULL;
  tree induc_val = NULL_TREE;
  tree adjustment_def = NULL;
  if (slp_node)
    ;
  else
    {
      /* Get at the scalar def before the loop, that defines the initial value
	 of the reduction variable.  */
      initial_def = PHI_ARG_DEF_FROM_EDGE (reduc_def_stmt,
					   loop_preheader_edge (loop));
      /* Optimize: for induction condition reduction, if we can't use zero
         for induc_val, use initial_def.  */
      if (STMT_VINFO_REDUC_TYPE (reduc_info) == INTEGER_INDUC_COND_REDUCTION)
	induc_val = STMT_VINFO_VEC_INDUC_COND_INITIAL_VAL (reduc_info);
      else if (double_reduc)
	;
      else if (nested_in_vect_loop)
	;
      else
	adjustment_def = STMT_VINFO_REDUC_EPILOGUE_ADJUSTMENT (reduc_info);
    }

  unsigned vec_num;
  int ncopies;
  if (slp_node)
    {
      vec_num = SLP_TREE_VEC_STMTS (slp_node_instance->reduc_phis).length ();
      ncopies = 1;
    }
  else
    {
      vec_num = 1;
      ncopies = 0;
      phi_info = STMT_VINFO_VEC_STMT (loop_vinfo->lookup_stmt (reduc_def_stmt));
      do
	{
	  ncopies++;
	  phi_info = STMT_VINFO_RELATED_STMT (phi_info);
	}
      while (phi_info);
    }

  /* For cond reductions we want to create a new vector (INDEX_COND_EXPR)
     which is updated with the current index of the loop for every match of
     the original loop's cond_expr (VEC_STMT).  This results in a vector
     containing the last time the condition passed for that vector lane.
     The first match will be a 1 to allow 0 to be used for non-matching
     indexes.  If there are no matches at all then the vector will be all
     zeroes.
   
     PR92772: This algorithm is broken for architectures that support
     masked vectors, but do not provide fold_extract_last.  */
  if (STMT_VINFO_REDUC_TYPE (reduc_info) == COND_REDUCTION)
    {
      auto_vec<std::pair<tree, bool>, 2> ccompares;
      stmt_vec_info cond_info = STMT_VINFO_REDUC_DEF (reduc_info);
      cond_info = vect_stmt_to_vectorize (cond_info);
      while (cond_info != reduc_info)
	{
	  if (gimple_assign_rhs_code (cond_info->stmt) == COND_EXPR)
	    {
	      gimple *vec_stmt = STMT_VINFO_VEC_STMT (cond_info)->stmt;
	      gcc_assert (gimple_assign_rhs_code (vec_stmt) == VEC_COND_EXPR);
	      ccompares.safe_push
		(std::make_pair (unshare_expr (gimple_assign_rhs1 (vec_stmt)),
				 STMT_VINFO_REDUC_IDX (cond_info) == 2));
	    }
	  cond_info
	    = loop_vinfo->lookup_def (gimple_op (cond_info->stmt,
						 1 + STMT_VINFO_REDUC_IDX
							(cond_info)));
	  cond_info = vect_stmt_to_vectorize (cond_info);
	}
      gcc_assert (ccompares.length () != 0);

      tree indx_before_incr, indx_after_incr;
      poly_uint64 nunits_out = TYPE_VECTOR_SUBPARTS (vectype);
      int scalar_precision
	= GET_MODE_PRECISION (SCALAR_TYPE_MODE (TREE_TYPE (vectype)));
      tree cr_index_scalar_type = make_unsigned_type (scalar_precision);
      tree cr_index_vector_type = get_related_vectype_for_scalar_type
	(TYPE_MODE (vectype), cr_index_scalar_type,
	 TYPE_VECTOR_SUBPARTS (vectype));

      /* First we create a simple vector induction variable which starts
	 with the values {1,2,3,...} (SERIES_VECT) and increments by the
	 vector size (STEP).  */

      /* Create a {1,2,3,...} vector.  */
      tree series_vect = build_index_vector (cr_index_vector_type, 1, 1);

      /* Create a vector of the step value.  */
      tree step = build_int_cst (cr_index_scalar_type, nunits_out);
      tree vec_step = build_vector_from_val (cr_index_vector_type, step);

      /* Create an induction variable.  */
      gimple_stmt_iterator incr_gsi;
      bool insert_after;
      standard_iv_increment_position (loop, &incr_gsi, &insert_after);
      create_iv (series_vect, vec_step, NULL_TREE, loop, &incr_gsi,
		 insert_after, &indx_before_incr, &indx_after_incr);

      /* Next create a new phi node vector (NEW_PHI_TREE) which starts
	 filled with zeros (VEC_ZERO).  */

      /* Create a vector of 0s.  */
      tree zero = build_zero_cst (cr_index_scalar_type);
      tree vec_zero = build_vector_from_val (cr_index_vector_type, zero);

      /* Create a vector phi node.  */
      tree new_phi_tree = make_ssa_name (cr_index_vector_type);
      new_phi = create_phi_node (new_phi_tree, loop->header);
      loop_vinfo->add_stmt (new_phi);
      add_phi_arg (as_a <gphi *> (new_phi), vec_zero,
		   loop_preheader_edge (loop), UNKNOWN_LOCATION);

      /* Now take the condition from the loops original cond_exprs
	 and produce a new cond_exprs (INDEX_COND_EXPR) which for
	 every match uses values from the induction variable
	 (INDEX_BEFORE_INCR) otherwise uses values from the phi node
	 (NEW_PHI_TREE).
	 Finally, we update the phi (NEW_PHI_TREE) to take the value of
	 the new cond_expr (INDEX_COND_EXPR).  */
      gimple_seq stmts = NULL;
      for (int i = ccompares.length () - 1; i != -1; --i)
	{
	  tree ccompare = ccompares[i].first;
	  if (ccompares[i].second)
	    new_phi_tree = gimple_build (&stmts, VEC_COND_EXPR,
					 cr_index_vector_type,
					 ccompare,
					 indx_before_incr, new_phi_tree);
	  else
	    new_phi_tree = gimple_build (&stmts, VEC_COND_EXPR,
					 cr_index_vector_type,
					 ccompare,
					 new_phi_tree, indx_before_incr);
	}
      gsi_insert_seq_before (&incr_gsi, stmts, GSI_SAME_STMT);
      stmt_vec_info index_vec_info
	= loop_vinfo->add_stmt (SSA_NAME_DEF_STMT (new_phi_tree));
      STMT_VINFO_VECTYPE (index_vec_info) = cr_index_vector_type;

      /* Update the phi with the vec cond.  */
      induction_index = new_phi_tree;
      add_phi_arg (as_a <gphi *> (new_phi), induction_index,
		   loop_latch_edge (loop), UNKNOWN_LOCATION);
    }

  /* 2. Create epilog code.
        The reduction epilog code operates across the elements of the vector
        of partial results computed by the vectorized loop.
        The reduction epilog code consists of:

        step 1: compute the scalar result in a vector (v_out2)
        step 2: extract the scalar result (s_out3) from the vector (v_out2)
        step 3: adjust the scalar result (s_out3) if needed.

        Step 1 can be accomplished using one the following three schemes:
          (scheme 1) using reduc_fn, if available.
          (scheme 2) using whole-vector shifts, if available.
          (scheme 3) using a scalar loop. In this case steps 1+2 above are
                     combined.

          The overall epilog code looks like this:

          s_out0 = phi <s_loop>         # original EXIT_PHI
          v_out1 = phi <VECT_DEF>       # NEW_EXIT_PHI
          v_out2 = reduce <v_out1>              # step 1
          s_out3 = extract_field <v_out2, 0>    # step 2
          s_out4 = adjust_result <s_out3>       # step 3

          (step 3 is optional, and steps 1 and 2 may be combined).
          Lastly, the uses of s_out0 are replaced by s_out4.  */


  /* 2.1 Create new loop-exit-phis to preserve loop-closed form:
         v_out1 = phi <VECT_DEF> 
         Store them in NEW_PHIS.  */
  if (double_reduc)
    loop = outer_loop;
  exit_bb = single_exit (loop)->dest;
  prev_phi_info = NULL;
  new_phis.create (slp_node ? vec_num : ncopies);
  for (unsigned i = 0; i < vec_num; i++)
    {
      if (slp_node)
	def = vect_get_slp_vect_def (slp_node, i);
      else
	def = gimple_get_lhs (STMT_VINFO_VEC_STMT (rdef_info)->stmt);
      for (j = 0; j < ncopies; j++)
        {
	  tree new_def = copy_ssa_name (def);
          phi = create_phi_node (new_def, exit_bb);
	  stmt_vec_info phi_info = loop_vinfo->add_stmt (phi);
          if (j == 0)
            new_phis.quick_push (phi);
          else
	    {
	      def = vect_get_vec_def_for_stmt_copy (loop_vinfo, def);
	      STMT_VINFO_RELATED_STMT (prev_phi_info) = phi_info;
	    }

          SET_PHI_ARG_DEF (phi, single_exit (loop)->dest_idx, def);
	  prev_phi_info = phi_info;
        }
    }

  exit_gsi = gsi_after_labels (exit_bb);

  /* 2.2 Get the relevant tree-code to use in the epilog for schemes 2,3
         (i.e. when reduc_fn is not available) and in the final adjustment
	 code (if needed).  Also get the original scalar reduction variable as
         defined in the loop.  In case STMT is a "pattern-stmt" (i.e. - it
         represents a reduction pattern), the tree-code and scalar-def are
         taken from the original stmt that the pattern-stmt (STMT) replaces.
         Otherwise (it is a regular reduction) - the tree-code and scalar-def
         are taken from STMT.  */

  stmt_vec_info orig_stmt_info = vect_orig_stmt (stmt_info);
  if (orig_stmt_info != stmt_info)
    {
      /* Reduction pattern  */
      gcc_assert (STMT_VINFO_IN_PATTERN_P (orig_stmt_info));
      gcc_assert (STMT_VINFO_RELATED_STMT (orig_stmt_info) == stmt_info);
    }
  
  scalar_dest = gimple_assign_lhs (orig_stmt_info->stmt);
  scalar_type = TREE_TYPE (scalar_dest);
  scalar_results.create (group_size); 
  new_scalar_dest = vect_create_destination_var (scalar_dest, NULL);
  bitsize = TYPE_SIZE (scalar_type);

  /* SLP reduction without reduction chain, e.g.,
     # a1 = phi <a2, a0>
     # b1 = phi <b2, b0>
     a2 = operation (a1)
     b2 = operation (b1)  */
  slp_reduc = (slp_node && !REDUC_GROUP_FIRST_ELEMENT (stmt_info));

  /* True if we should implement SLP_REDUC using native reduction operations
     instead of scalar operations.  */
  direct_slp_reduc = (reduc_fn != IFN_LAST
		      && slp_reduc
		      && !TYPE_VECTOR_SUBPARTS (vectype).is_constant ());

  /* In case of reduction chain, e.g.,
     # a1 = phi <a3, a0>
     a2 = operation (a1)
     a3 = operation (a2),

     we may end up with more than one vector result.  Here we reduce them to
     one vector.  */
  if (REDUC_GROUP_FIRST_ELEMENT (stmt_info) || direct_slp_reduc)
    {
      gimple_seq stmts = NULL;
      tree first_vect = PHI_RESULT (new_phis[0]);
      first_vect = gimple_convert (&stmts, vectype, first_vect);
      for (k = 1; k < new_phis.length (); k++)
        {
	  gimple *next_phi = new_phis[k];
          tree second_vect = PHI_RESULT (next_phi);
	  second_vect = gimple_convert (&stmts, vectype, second_vect);
          first_vect = gimple_build (&stmts, code, vectype,
				     first_vect, second_vect);
        }
      gsi_insert_seq_before (&exit_gsi, stmts, GSI_SAME_STMT);

      new_phi_result = first_vect;
      new_phis.truncate (0);
      new_phis.safe_push (SSA_NAME_DEF_STMT (first_vect));
    }
  /* Likewise if we couldn't use a single defuse cycle.  */
  else if (ncopies > 1)
    {
      gcc_assert (new_phis.length () == 1);
      gimple_seq stmts = NULL;
      tree first_vect = PHI_RESULT (new_phis[0]);
      first_vect = gimple_convert (&stmts, vectype, first_vect);
      stmt_vec_info next_phi_info = loop_vinfo->lookup_stmt (new_phis[0]);
      for (int k = 1; k < ncopies; ++k)
	{
	  next_phi_info = STMT_VINFO_RELATED_STMT (next_phi_info);
	  tree second_vect = PHI_RESULT (next_phi_info->stmt);
	  second_vect = gimple_convert (&stmts, vectype, second_vect);
	  first_vect = gimple_build (&stmts, code, vectype,
				     first_vect, second_vect);
	}
      gsi_insert_seq_before (&exit_gsi, stmts, GSI_SAME_STMT);
      new_phi_result = first_vect;
      new_phis.truncate (0);
      new_phis.safe_push (SSA_NAME_DEF_STMT (first_vect));
    }
  else
    new_phi_result = PHI_RESULT (new_phis[0]);

  if (STMT_VINFO_REDUC_TYPE (reduc_info) == COND_REDUCTION
      && reduc_fn != IFN_LAST)
    {
      /* For condition reductions, we have a vector (NEW_PHI_RESULT) containing
	 various data values where the condition matched and another vector
	 (INDUCTION_INDEX) containing all the indexes of those matches.  We
	 need to extract the last matching index (which will be the index with
	 highest value) and use this to index into the data vector.
	 For the case where there were no matches, the data vector will contain
	 all default values and the index vector will be all zeros.  */

      /* Get various versions of the type of the vector of indexes.  */
      tree index_vec_type = TREE_TYPE (induction_index);
      gcc_checking_assert (TYPE_UNSIGNED (index_vec_type));
      tree index_scalar_type = TREE_TYPE (index_vec_type);
      tree index_vec_cmp_type = truth_type_for (index_vec_type);

      /* Get an unsigned integer version of the type of the data vector.  */
      int scalar_precision
	= GET_MODE_PRECISION (SCALAR_TYPE_MODE (scalar_type));
      tree scalar_type_unsigned = make_unsigned_type (scalar_precision);
      tree vectype_unsigned = build_vector_type
	(scalar_type_unsigned, TYPE_VECTOR_SUBPARTS (vectype));

      /* First we need to create a vector (ZERO_VEC) of zeros and another
	 vector (MAX_INDEX_VEC) filled with the last matching index, which we
	 can create using a MAX reduction and then expanding.
	 In the case where the loop never made any matches, the max index will
	 be zero.  */

      /* Vector of {0, 0, 0,...}.  */
      tree zero_vec = build_zero_cst (vectype);

      gimple_seq stmts = NULL;
      new_phi_result = gimple_convert (&stmts, vectype, new_phi_result);
      gsi_insert_seq_before (&exit_gsi, stmts, GSI_SAME_STMT);

      /* Find maximum value from the vector of found indexes.  */
      tree max_index = make_ssa_name (index_scalar_type);
      gcall *max_index_stmt = gimple_build_call_internal (IFN_REDUC_MAX,
							  1, induction_index);
      gimple_call_set_lhs (max_index_stmt, max_index);
      gsi_insert_before (&exit_gsi, max_index_stmt, GSI_SAME_STMT);

      /* Vector of {max_index, max_index, max_index,...}.  */
      tree max_index_vec = make_ssa_name (index_vec_type);
      tree max_index_vec_rhs = build_vector_from_val (index_vec_type,
						      max_index);
      gimple *max_index_vec_stmt = gimple_build_assign (max_index_vec,
							max_index_vec_rhs);
      gsi_insert_before (&exit_gsi, max_index_vec_stmt, GSI_SAME_STMT);

      /* Next we compare the new vector (MAX_INDEX_VEC) full of max indexes
	 with the vector (INDUCTION_INDEX) of found indexes, choosing values
	 from the data vector (NEW_PHI_RESULT) for matches, 0 (ZERO_VEC)
	 otherwise.  Only one value should match, resulting in a vector
	 (VEC_COND) with one data value and the rest zeros.
	 In the case where the loop never made any matches, every index will
	 match, resulting in a vector with all data values (which will all be
	 the default value).  */

      /* Compare the max index vector to the vector of found indexes to find
	 the position of the max value.  */
      tree vec_compare = make_ssa_name (index_vec_cmp_type);
      gimple *vec_compare_stmt = gimple_build_assign (vec_compare, EQ_EXPR,
						      induction_index,
						      max_index_vec);
      gsi_insert_before (&exit_gsi, vec_compare_stmt, GSI_SAME_STMT);

      /* Use the compare to choose either values from the data vector or
	 zero.  */
      tree vec_cond = make_ssa_name (vectype);
      gimple *vec_cond_stmt = gimple_build_assign (vec_cond, VEC_COND_EXPR,
						   vec_compare, new_phi_result,
						   zero_vec);
      gsi_insert_before (&exit_gsi, vec_cond_stmt, GSI_SAME_STMT);

      /* Finally we need to extract the data value from the vector (VEC_COND)
	 into a scalar (MATCHED_DATA_REDUC).  Logically we want to do a OR
	 reduction, but because this doesn't exist, we can use a MAX reduction
	 instead.  The data value might be signed or a float so we need to cast
	 it first.
	 In the case where the loop never made any matches, the data values are
	 all identical, and so will reduce down correctly.  */

      /* Make the matched data values unsigned.  */
      tree vec_cond_cast = make_ssa_name (vectype_unsigned);
      tree vec_cond_cast_rhs = build1 (VIEW_CONVERT_EXPR, vectype_unsigned,
				       vec_cond);
      gimple *vec_cond_cast_stmt = gimple_build_assign (vec_cond_cast,
							VIEW_CONVERT_EXPR,
							vec_cond_cast_rhs);
      gsi_insert_before (&exit_gsi, vec_cond_cast_stmt, GSI_SAME_STMT);

      /* Reduce down to a scalar value.  */
      tree data_reduc = make_ssa_name (scalar_type_unsigned);
      gcall *data_reduc_stmt = gimple_build_call_internal (IFN_REDUC_MAX,
							   1, vec_cond_cast);
      gimple_call_set_lhs (data_reduc_stmt, data_reduc);
      gsi_insert_before (&exit_gsi, data_reduc_stmt, GSI_SAME_STMT);

      /* Convert the reduced value back to the result type and set as the
	 result.  */
      stmts = NULL;
      new_temp = gimple_build (&stmts, VIEW_CONVERT_EXPR, scalar_type,
			       data_reduc);
      gsi_insert_seq_before (&exit_gsi, stmts, GSI_SAME_STMT);
      scalar_results.safe_push (new_temp);
    }
  else if (STMT_VINFO_REDUC_TYPE (reduc_info) == COND_REDUCTION
	   && reduc_fn == IFN_LAST)
    {
      /* Condition reduction without supported IFN_REDUC_MAX.  Generate
	 idx = 0;
         idx_val = induction_index[0];
	 val = data_reduc[0];
         for (idx = 0, val = init, i = 0; i < nelts; ++i)
	   if (induction_index[i] > idx_val)
	     val = data_reduc[i], idx_val = induction_index[i];
	 return val;  */

      tree data_eltype = TREE_TYPE (TREE_TYPE (new_phi_result));
      tree idx_eltype = TREE_TYPE (TREE_TYPE (induction_index));
      unsigned HOST_WIDE_INT el_size = tree_to_uhwi (TYPE_SIZE (idx_eltype));
      poly_uint64 nunits = TYPE_VECTOR_SUBPARTS (TREE_TYPE (induction_index));
      /* Enforced by vectorizable_reduction, which ensures we have target
	 support before allowing a conditional reduction on variable-length
	 vectors.  */
      unsigned HOST_WIDE_INT v_size = el_size * nunits.to_constant ();
      tree idx_val = NULL_TREE, val = NULL_TREE;
      for (unsigned HOST_WIDE_INT off = 0; off < v_size; off += el_size)
	{
	  tree old_idx_val = idx_val;
	  tree old_val = val;
	  idx_val = make_ssa_name (idx_eltype);
	  epilog_stmt = gimple_build_assign (idx_val, BIT_FIELD_REF,
					     build3 (BIT_FIELD_REF, idx_eltype,
						     induction_index,
						     bitsize_int (el_size),
						     bitsize_int (off)));
	  gsi_insert_before (&exit_gsi, epilog_stmt, GSI_SAME_STMT);
	  val = make_ssa_name (data_eltype);
	  epilog_stmt = gimple_build_assign (val, BIT_FIELD_REF,
					     build3 (BIT_FIELD_REF,
						     data_eltype,
						     new_phi_result,
						     bitsize_int (el_size),
						     bitsize_int (off)));
	  gsi_insert_before (&exit_gsi, epilog_stmt, GSI_SAME_STMT);
	  if (off != 0)
	    {
	      tree new_idx_val = idx_val;
	      if (off != v_size - el_size)
		{
		  new_idx_val = make_ssa_name (idx_eltype);
		  epilog_stmt = gimple_build_assign (new_idx_val,
						     MAX_EXPR, idx_val,
						     old_idx_val);
		  gsi_insert_before (&exit_gsi, epilog_stmt, GSI_SAME_STMT);
		}
	      tree new_val = make_ssa_name (data_eltype);
	      epilog_stmt = gimple_build_assign (new_val,
						 COND_EXPR,
						 build2 (GT_EXPR,
							 boolean_type_node,
							 idx_val,
							 old_idx_val),
						 val, old_val);
	      gsi_insert_before (&exit_gsi, epilog_stmt, GSI_SAME_STMT);
	      idx_val = new_idx_val;
	      val = new_val;
	    }
	}
      /* Convert the reduced value back to the result type and set as the
	 result.  */
      gimple_seq stmts = NULL;