Training courses

Kernel and Embedded Linux

Bootlin training courses

Embedded Linux, kernel,
Yocto Project, Buildroot, real-time,
graphics, boot time, debugging...

Bootlin logo

Elixir Cross Referencer

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
/* Utilities for ipa analysis.
   Copyright (C) 2005-2017 Free Software Foundation, Inc.
   Contributed by Kenneth Zadeck <zadeck@naturalbridge.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 "tree.h"
#include "gimple.h"
#include "predict.h"
#include "alloc-pool.h"
#include "cgraph.h"
#include "lto-streamer.h"
#include "dumpfile.h"
#include "splay-tree.h"
#include "ipa-utils.h"
#include "symbol-summary.h"
#include "tree-vrp.h"
#include "ipa-prop.h"
#include "ipa-inline.h"

/* Debugging function for postorder and inorder code. NOTE is a string
   that is printed before the nodes are printed.  ORDER is an array of
   cgraph_nodes that has COUNT useful nodes in it.  */

void
ipa_print_order (FILE* out,
		 const char * note,
		 struct cgraph_node** order,
		 int count)
{
  int i;
  fprintf (out, "\n\n ordered call graph: %s\n", note);

  for (i = count - 1; i >= 0; i--)
    order[i]->dump (out);
  fprintf (out, "\n");
  fflush (out);
}


struct searchc_env {
  struct cgraph_node **stack;
  struct cgraph_node **result;
  int stack_size;
  int order_pos;
  splay_tree nodes_marked_new;
  bool reduce;
  int count;
};

/* This is an implementation of Tarjan's strongly connected region
   finder as reprinted in Aho Hopcraft and Ullman's The Design and
   Analysis of Computer Programs (1975) pages 192-193.  This version
   has been customized for cgraph_nodes.  The env parameter is because
   it is recursive and there are no nested functions here.  This
   function should only be called from itself or
   ipa_reduced_postorder.  ENV is a stack env and would be
   unnecessary if C had nested functions.  V is the node to start
   searching from.  */

static void
searchc (struct searchc_env* env, struct cgraph_node *v,
	 bool (*ignore_edge) (struct cgraph_edge *))
{
  struct cgraph_edge *edge;
  struct ipa_dfs_info *v_info = (struct ipa_dfs_info *) v->aux;

  /* mark node as old */
  v_info->new_node = false;
  splay_tree_remove (env->nodes_marked_new, v->uid);

  v_info->dfn_number = env->count;
  v_info->low_link = env->count;
  env->count++;
  env->stack[(env->stack_size)++] = v;
  v_info->on_stack = true;

  for (edge = v->callees; edge; edge = edge->next_callee)
    {
      struct ipa_dfs_info * w_info;
      enum availability avail;
      struct cgraph_node *w = edge->callee->ultimate_alias_target (&avail);

      if (!w || (ignore_edge && ignore_edge (edge)))
        continue;

      if (w->aux
	  && (avail > AVAIL_INTERPOSABLE
	      || avail == AVAIL_INTERPOSABLE))
	{
	  w_info = (struct ipa_dfs_info *) w->aux;
	  if (w_info->new_node)
	    {
	      searchc (env, w, ignore_edge);
	      v_info->low_link =
		(v_info->low_link < w_info->low_link) ?
		v_info->low_link : w_info->low_link;
	    }
	  else
	    if ((w_info->dfn_number < v_info->dfn_number)
		&& (w_info->on_stack))
	      v_info->low_link =
		(w_info->dfn_number < v_info->low_link) ?
		w_info->dfn_number : v_info->low_link;
	}
    }


  if (v_info->low_link == v_info->dfn_number)
    {
      struct cgraph_node *last = NULL;
      struct cgraph_node *x;
      struct ipa_dfs_info *x_info;
      do {
	x = env->stack[--(env->stack_size)];
	x_info = (struct ipa_dfs_info *) x->aux;
	x_info->on_stack = false;
	x_info->scc_no = v_info->dfn_number;

	if (env->reduce)
	  {
	    x_info->next_cycle = last;
	    last = x;
	  }
	else
	  env->result[env->order_pos++] = x;
      }
      while (v != x);
      if (env->reduce)
	env->result[env->order_pos++] = v;
    }
}

/* Topsort the call graph by caller relation.  Put the result in ORDER.

   The REDUCE flag is true if you want the cycles reduced to single nodes.
   You can use ipa_get_nodes_in_cycle to obtain a vector containing all real
   call graph nodes in a reduced node.

   Set ALLOW_OVERWRITABLE if nodes with such availability should be included.
   IGNORE_EDGE, if non-NULL is a hook that may make some edges insignificant
   for the topological sort.   */

int
ipa_reduced_postorder (struct cgraph_node **order,
		       bool reduce,
		       bool (*ignore_edge) (struct cgraph_edge *))
{
  struct cgraph_node *node;
  struct searchc_env env;
  splay_tree_node result;
  env.stack = XCNEWVEC (struct cgraph_node *, symtab->cgraph_count);
  env.stack_size = 0;
  env.result = order;
  env.order_pos = 0;
  env.nodes_marked_new = splay_tree_new (splay_tree_compare_ints, 0, 0);
  env.count = 1;
  env.reduce = reduce;

  FOR_EACH_DEFINED_FUNCTION (node)
    {
      enum availability avail = node->get_availability ();

      if (avail > AVAIL_INTERPOSABLE
	  || avail == AVAIL_INTERPOSABLE)
	{
	  /* Reuse the info if it is already there.  */
	  struct ipa_dfs_info *info = (struct ipa_dfs_info *) node->aux;
	  if (!info)
	    info = XCNEW (struct ipa_dfs_info);
	  info->new_node = true;
	  info->on_stack = false;
	  info->next_cycle = NULL;
	  node->aux = info;

	  splay_tree_insert (env.nodes_marked_new,
			     (splay_tree_key)node->uid,
			     (splay_tree_value)node);
	}
      else
	node->aux = NULL;
    }
  result = splay_tree_min (env.nodes_marked_new);
  while (result)
    {
      node = (struct cgraph_node *)result->value;
      searchc (&env, node, ignore_edge);
      result = splay_tree_min (env.nodes_marked_new);
    }
  splay_tree_delete (env.nodes_marked_new);
  free (env.stack);

  return env.order_pos;
}

/* Deallocate all ipa_dfs_info structures pointed to by the aux pointer of call
   graph nodes.  */

void
ipa_free_postorder_info (void)
{
  struct cgraph_node *node;
  FOR_EACH_DEFINED_FUNCTION (node)
    {
      /* Get rid of the aux information.  */
      if (node->aux)
	{
	  free (node->aux);
	  node->aux = NULL;
	}
    }
}

/* Get the set of nodes for the cycle in the reduced call graph starting
   from NODE.  */

vec<cgraph_node *>
ipa_get_nodes_in_cycle (struct cgraph_node *node)
{
  vec<cgraph_node *> v = vNULL;
  struct ipa_dfs_info *node_dfs_info;
  while (node)
    {
      v.safe_push (node);
      node_dfs_info = (struct ipa_dfs_info *) node->aux;
      node = node_dfs_info->next_cycle;
    }
  return v;
}

/* Return true iff the CS is an edge within a strongly connected component as
   computed by ipa_reduced_postorder.  */

bool
ipa_edge_within_scc (struct cgraph_edge *cs)
{
  struct ipa_dfs_info *caller_dfs = (struct ipa_dfs_info *) cs->caller->aux;
  struct ipa_dfs_info *callee_dfs;
  struct cgraph_node *callee = cs->callee->function_symbol ();

  callee_dfs = (struct ipa_dfs_info *) callee->aux;
  return (caller_dfs
	  && callee_dfs
	  && caller_dfs->scc_no == callee_dfs->scc_no);
}

struct postorder_stack
{
  struct cgraph_node *node;
  struct cgraph_edge *edge;
  int ref;
};

/* Fill array order with all nodes with output flag set in the reverse
   topological order.  Return the number of elements in the array.
   FIXME: While walking, consider aliases, too.  */

int
ipa_reverse_postorder (struct cgraph_node **order)
{
  struct cgraph_node *node, *node2;
  int stack_size = 0;
  int order_pos = 0;
  struct cgraph_edge *edge;
  int pass;
  struct ipa_ref *ref = NULL;

  struct postorder_stack *stack =
    XCNEWVEC (struct postorder_stack, symtab->cgraph_count);

  /* We have to deal with cycles nicely, so use a depth first traversal
     output algorithm.  Ignore the fact that some functions won't need
     to be output and put them into order as well, so we get dependencies
     right through inline functions.  */
  FOR_EACH_FUNCTION (node)
    node->aux = NULL;
  for (pass = 0; pass < 2; pass++)
    FOR_EACH_FUNCTION (node)
      if (!node->aux
	  && (pass
	      || (!node->address_taken
		  && !node->global.inlined_to
		  && !node->alias && !node->thunk.thunk_p
		  && !node->only_called_directly_p ())))
	{
	  stack_size = 0;
          stack[stack_size].node = node;
	  stack[stack_size].edge = node->callers;
	  stack[stack_size].ref = 0;
	  node->aux = (void *)(size_t)1;
	  while (stack_size >= 0)
	    {
	      while (true)
		{
		  node2 = NULL;
		  while (stack[stack_size].edge && !node2)
		    {
		      edge = stack[stack_size].edge;
		      node2 = edge->caller;
		      stack[stack_size].edge = edge->next_caller;
		      /* Break possible cycles involving always-inline
			 functions by ignoring edges from always-inline
			 functions to non-always-inline functions.  */
		      if (DECL_DISREGARD_INLINE_LIMITS (edge->caller->decl)
			  && !DECL_DISREGARD_INLINE_LIMITS
			    (edge->callee->function_symbol ()->decl))
			node2 = NULL;
		    }
		  for (; stack[stack_size].node->iterate_referring (
						       stack[stack_size].ref,
						       ref) && !node2;
		       stack[stack_size].ref++)
		    {
		      if (ref->use == IPA_REF_ALIAS)
			node2 = dyn_cast <cgraph_node *> (ref->referring);
		    }
		  if (!node2)
		    break;
		  if (!node2->aux)
		    {
		      stack[++stack_size].node = node2;
		      stack[stack_size].edge = node2->callers;
		      stack[stack_size].ref = 0;
		      node2->aux = (void *)(size_t)1;
		    }
		}
	      order[order_pos++] = stack[stack_size--].node;
	    }
	}
  free (stack);
  FOR_EACH_FUNCTION (node)
    node->aux = NULL;
  return order_pos;
}



/* Given a memory reference T, will return the variable at the bottom
   of the access.  Unlike get_base_address, this will recurse through
   INDIRECT_REFS.  */

tree
get_base_var (tree t)
{
  while (!SSA_VAR_P (t)
	 && (!CONSTANT_CLASS_P (t))
	 && TREE_CODE (t) != LABEL_DECL
	 && TREE_CODE (t) != FUNCTION_DECL
	 && TREE_CODE (t) != CONST_DECL
	 && TREE_CODE (t) != CONSTRUCTOR)
    {
      t = TREE_OPERAND (t, 0);
    }
  return t;
}


/* SRC and DST are going to be merged.  Take SRC's profile and merge it into
   DST so it is not going to be lost.  Possibly destroy SRC's body on the way
   unless PRESERVE_BODY is set.  */

void
ipa_merge_profiles (struct cgraph_node *dst,
		    struct cgraph_node *src,
		    bool preserve_body)
{
  tree oldsrcdecl = src->decl;
  struct function *srccfun, *dstcfun;
  bool match = true;

  if (!src->definition
      || !dst->definition)
    return;
  if (src->frequency < dst->frequency)
    src->frequency = dst->frequency;

  /* Time profiles are merged.  */
  if (dst->tp_first_run > src->tp_first_run && src->tp_first_run)
    dst->tp_first_run = src->tp_first_run;

  if (src->profile_id && !dst->profile_id)
    dst->profile_id = src->profile_id;

  if (!dst->count)
    return;
  if (!src->count || src->alias)
    return;
  if (symtab->dump_file)
    {
      fprintf (symtab->dump_file, "Merging profiles of %s/%i to %s/%i\n",
	       xstrdup_for_dump (src->name ()), src->order,
	       xstrdup_for_dump (dst->name ()), dst->order);
    }
  dst->count += src->count;

  /* This is ugly.  We need to get both function bodies into memory.
     If declaration is merged, we need to duplicate it to be able
     to load body that is being replaced.  This makes symbol table
     temporarily inconsistent.  */
  if (src->decl == dst->decl)
    {
      struct lto_in_decl_state temp;
      struct lto_in_decl_state *state;

      /* We are going to move the decl, we want to remove its file decl data.
	 and link these with the new decl. */
      temp.fn_decl = src->decl;
      lto_in_decl_state **slot
	= src->lto_file_data->function_decl_states->find_slot (&temp,
							       NO_INSERT);
      state = *slot;
      src->lto_file_data->function_decl_states->clear_slot (slot);
      gcc_assert (state);

      /* Duplicate the decl and be sure it does not link into body of DST.  */
      src->decl = copy_node (src->decl);
      DECL_STRUCT_FUNCTION (src->decl) = NULL;
      DECL_ARGUMENTS (src->decl) = NULL;
      DECL_INITIAL (src->decl) = NULL;
      DECL_RESULT (src->decl) = NULL;

      /* Associate the decl state with new declaration, so LTO streamer
 	 can look it up.  */
      state->fn_decl = src->decl;
      slot
	= src->lto_file_data->function_decl_states->find_slot (state, INSERT);
      gcc_assert (!*slot);
      *slot = state;
    }
  src->get_untransformed_body ();
  dst->get_untransformed_body ();
  srccfun = DECL_STRUCT_FUNCTION (src->decl);
  dstcfun = DECL_STRUCT_FUNCTION (dst->decl);
  if (n_basic_blocks_for_fn (srccfun)
      != n_basic_blocks_for_fn (dstcfun))
    {
      if (symtab->dump_file)
	fprintf (symtab->dump_file,
		 "Giving up; number of basic block mismatch.\n");
      match = false;
    }
  else if (last_basic_block_for_fn (srccfun)
	   != last_basic_block_for_fn (dstcfun))
    {
      if (symtab->dump_file)
	fprintf (symtab->dump_file,
		 "Giving up; last block mismatch.\n");
      match = false;
    }
  else 
    {
      basic_block srcbb, dstbb;

      FOR_ALL_BB_FN (srcbb, srccfun)
	{
	  unsigned int i;

	  dstbb = BASIC_BLOCK_FOR_FN (dstcfun, srcbb->index);
	  if (dstbb == NULL)
	    {
	      if (symtab->dump_file)
		fprintf (symtab->dump_file,
			 "No matching block for bb %i.\n",
			 srcbb->index);
	      match = false;
	      break;
	    }
	  if (EDGE_COUNT (srcbb->succs) != EDGE_COUNT (dstbb->succs))
	    {
	      if (symtab->dump_file)
		fprintf (symtab->dump_file,
			 "Edge count mistmatch for bb %i.\n",
			 srcbb->index);
	      match = false;
	      break;
	    }
	  for (i = 0; i < EDGE_COUNT (srcbb->succs); i++)
	    {
	      edge srce = EDGE_SUCC (srcbb, i);
	      edge dste = EDGE_SUCC (dstbb, i);
	      if (srce->dest->index != dste->dest->index)
		{
		  if (symtab->dump_file)
		    fprintf (symtab->dump_file,
			     "Succ edge mistmatch for bb %i.\n",
			     srce->dest->index);
		  match = false;
		  break;
		}
	    }
	}
    }
  if (match)
    {
      struct cgraph_edge *e, *e2;
      basic_block srcbb, dstbb;

      /* TODO: merge also statement histograms.  */
      FOR_ALL_BB_FN (srcbb, srccfun)
	{
	  unsigned int i;

	  dstbb = BASIC_BLOCK_FOR_FN (dstcfun, srcbb->index);
	  dstbb->count += srcbb->count;
	  for (i = 0; i < EDGE_COUNT (srcbb->succs); i++)
	    {
	      edge srce = EDGE_SUCC (srcbb, i);
	      edge dste = EDGE_SUCC (dstbb, i);
	      dste->count += srce->count;
	    }
	}
      push_cfun (dstcfun);
      counts_to_freqs ();
      compute_function_frequency ();
      pop_cfun ();
      for (e = dst->callees; e; e = e->next_callee)
	{
	  if (e->speculative)
	    continue;
	  e->count = gimple_bb (e->call_stmt)->count;
	  e->frequency = compute_call_stmt_bb_frequency
			     (dst->decl,
			      gimple_bb (e->call_stmt));
	}
      for (e = dst->indirect_calls, e2 = src->indirect_calls; e;
	   e2 = (e2 ? e2->next_callee : NULL), e = e->next_callee)
	{
	  gcov_type count = gimple_bb (e->call_stmt)->count;
	  int freq = compute_call_stmt_bb_frequency
			(dst->decl,
			 gimple_bb (e->call_stmt));
	  /* When call is speculative, we need to re-distribute probabilities
	     the same way as they was.  This is not really correct because
	     in the other copy the speculation may differ; but probably it
	     is not really worth the effort.  */
	  if (e->speculative)
	    {
	      cgraph_edge *direct, *indirect;
	      cgraph_edge *direct2 = NULL, *indirect2 = NULL;
	      ipa_ref *ref;

	      e->speculative_call_info (direct, indirect, ref);
	      gcc_assert (e == indirect);
	      if (e2 && e2->speculative)
	        e2->speculative_call_info (direct2, indirect2, ref);
	      if (indirect->count || direct->count)
		{
		  /* We should mismatch earlier if there is no matching
		     indirect edge.  */
		  if (!e2)
		    {
		      if (dump_file)
		        fprintf (dump_file,
				 "Mismatch in merging indirect edges\n");
		    }
		  else if (!e2->speculative)
		    indirect->count += e2->count;
		  else if (e2->speculative)
		    {
		      if (DECL_ASSEMBLER_NAME (direct2->callee->decl)
			  != DECL_ASSEMBLER_NAME (direct->callee->decl))
			{
			  if (direct2->count >= direct->count)
			    {
			      direct->redirect_callee (direct2->callee);
			      indirect->count += indirect2->count
						 + direct->count;
			      direct->count = direct2->count;
			    }
			  else
			    indirect->count += indirect2->count + direct2->count;
			}
		      else
			{
			   direct->count += direct2->count;
			   indirect->count += indirect2->count;
			}
		    }
		  int  prob = RDIV (direct->count * REG_BR_PROB_BASE ,
				    direct->count + indirect->count);
		  direct->frequency = RDIV (freq * prob, REG_BR_PROB_BASE);
		  indirect->frequency = RDIV (freq * (REG_BR_PROB_BASE - prob),
					      REG_BR_PROB_BASE);
		}
	      else
		/* At the moment we should have only profile feedback based
		   speculations when merging.  */
		gcc_unreachable ();
	    }
	  else if (e2 && e2->speculative)
	    {
	      cgraph_edge *direct, *indirect;
	      ipa_ref *ref;

	      e2->speculative_call_info (direct, indirect, ref);
	      e->count = count;
	      e->frequency = freq;
	      int prob = RDIV (direct->count * REG_BR_PROB_BASE, e->count);
	      e->make_speculative (direct->callee, direct->count,
				   RDIV (freq * prob, REG_BR_PROB_BASE));
	    }
	  else
	    {
	      e->count = count;
	      e->frequency = freq;
	    }
	}
      if (!preserve_body)
        src->release_body ();
      inline_update_overall_summary (dst);
    }
  /* TODO: if there is no match, we can scale up.  */
  src->decl = oldsrcdecl;
}

/* Return true if call to DEST is known to be self-recusive call withing FUNC.   */

bool
recursive_call_p (tree func, tree dest)
{
  struct cgraph_node *dest_node = cgraph_node::get_create (dest);
  struct cgraph_node *cnode = cgraph_node::get_create (func);
  ipa_ref *alias;
  enum availability avail;

  gcc_assert (!cnode->alias);
  if (cnode != dest_node->ultimate_alias_target (&avail))
    return false;
  if (avail >= AVAIL_AVAILABLE)
    return true;
  if (!dest_node->semantically_equivalent_p (cnode))
    return false;
  /* If there is only one way to call the fuction or we know all of them
     are semantically equivalent, we still can consider call recursive.  */
  FOR_EACH_ALIAS (cnode, alias)
    if (!dest_node->semantically_equivalent_p (alias->referring))
      return false;
  return true;
}