1 /* Code sinking for trees
2 Copyright (C) 2001-2021 Free Software Foundation, Inc.
3 Contributed by Daniel Berlin <dan@dberlin.org>
4
5 This file is part of GCC.
6
7 GCC is free software; you can redistribute it and/or modify
8 it under the terms of the GNU General Public License as published by
9 the Free Software Foundation; either version 3, or (at your option)
10 any later version.
11
12 GCC is distributed in the hope that it will be useful,
13 but WITHOUT ANY WARRANTY; without even the implied warranty of
14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 GNU General Public License for more details.
16
17 You should have received a copy of the GNU General Public License
18 along with GCC; see the file COPYING3. If not see
19 <http://www.gnu.org/licenses/>. */
20
21 #include "config.h"
22 #include "system.h"
23 #include "coretypes.h"
24 #include "backend.h"
25 #include "tree.h"
26 #include "gimple.h"
27 #include "cfghooks.h"
28 #include "tree-pass.h"
29 #include "ssa.h"
30 #include "gimple-pretty-print.h"
31 #include "fold-const.h"
32 #include "stor-layout.h"
33 #include "cfganal.h"
34 #include "gimple-iterator.h"
35 #include "tree-cfg.h"
36 #include "cfgloop.h"
37 #include "tree-eh.h"
38
39 /* TODO:
40 1. Sinking store only using scalar promotion (IE without moving the RHS):
41
42 *q = p;
43 p = p + 1;
44 if (something)
45 *q = <not p>;
46 else
47 y = *q;
48
49
50 should become
51 sinktemp = p;
52 p = p + 1;
53 if (something)
54 *q = <not p>;
55 else
56 {
57 *q = sinktemp;
58 y = *q
59 }
60 Store copy propagation will take care of the store elimination above.
61
62
63 2. Sinking using Partial Dead Code Elimination. */
64
65
66 static struct
67 {
68 /* The number of statements sunk down the flowgraph by code sinking. */
69 int sunk;
70
71 /* The number of stores commoned and sunk down by store commoning. */
72 int commoned;
73 } sink_stats;
74
75
76 /* Given a PHI, and one of its arguments (DEF), find the edge for
77 that argument and return it. If the argument occurs twice in the PHI node,
78 we return NULL. */
79
80 static basic_block
find_bb_for_arg(gphi * phi,tree def)81 find_bb_for_arg (gphi *phi, tree def)
82 {
83 size_t i;
84 bool foundone = false;
85 basic_block result = NULL;
86 for (i = 0; i < gimple_phi_num_args (phi); i++)
87 if (PHI_ARG_DEF (phi, i) == def)
88 {
89 if (foundone)
90 return NULL;
91 foundone = true;
92 result = gimple_phi_arg_edge (phi, i)->src;
93 }
94 return result;
95 }
96
97 /* When the first immediate use is in a statement, then return true if all
98 immediate uses in IMM are in the same statement.
99 We could also do the case where the first immediate use is in a phi node,
100 and all the other uses are in phis in the same basic block, but this
101 requires some expensive checking later (you have to make sure no def/vdef
102 in the statement occurs for multiple edges in the various phi nodes it's
103 used in, so that you only have one place you can sink it to. */
104
105 static bool
all_immediate_uses_same_place(def_operand_p def_p)106 all_immediate_uses_same_place (def_operand_p def_p)
107 {
108 tree var = DEF_FROM_PTR (def_p);
109 imm_use_iterator imm_iter;
110 use_operand_p use_p;
111
112 gimple *firstuse = NULL;
113 FOR_EACH_IMM_USE_FAST (use_p, imm_iter, var)
114 {
115 if (is_gimple_debug (USE_STMT (use_p)))
116 continue;
117 if (firstuse == NULL)
118 firstuse = USE_STMT (use_p);
119 else
120 if (firstuse != USE_STMT (use_p))
121 return false;
122 }
123
124 return true;
125 }
126
127 /* Find the nearest common dominator of all of the immediate uses in IMM. */
128
129 static basic_block
nearest_common_dominator_of_uses(def_operand_p def_p,bool * debug_stmts)130 nearest_common_dominator_of_uses (def_operand_p def_p, bool *debug_stmts)
131 {
132 tree var = DEF_FROM_PTR (def_p);
133 auto_bitmap blocks;
134 basic_block commondom;
135 unsigned int j;
136 bitmap_iterator bi;
137 imm_use_iterator imm_iter;
138 use_operand_p use_p;
139
140 FOR_EACH_IMM_USE_FAST (use_p, imm_iter, var)
141 {
142 gimple *usestmt = USE_STMT (use_p);
143 basic_block useblock;
144
145 if (gphi *phi = dyn_cast <gphi *> (usestmt))
146 {
147 int idx = PHI_ARG_INDEX_FROM_USE (use_p);
148
149 useblock = gimple_phi_arg_edge (phi, idx)->src;
150 }
151 else if (is_gimple_debug (usestmt))
152 {
153 *debug_stmts = true;
154 continue;
155 }
156 else
157 {
158 useblock = gimple_bb (usestmt);
159 }
160
161 /* Short circuit. Nothing dominates the entry block. */
162 if (useblock == ENTRY_BLOCK_PTR_FOR_FN (cfun))
163 return NULL;
164
165 bitmap_set_bit (blocks, useblock->index);
166 }
167 commondom = BASIC_BLOCK_FOR_FN (cfun, bitmap_first_set_bit (blocks));
168 EXECUTE_IF_SET_IN_BITMAP (blocks, 0, j, bi)
169 commondom = nearest_common_dominator (CDI_DOMINATORS, commondom,
170 BASIC_BLOCK_FOR_FN (cfun, j));
171 return commondom;
172 }
173
174 /* Given EARLY_BB and LATE_BB, two blocks in a path through the dominator
175 tree, return the best basic block between them (inclusive) to place
176 statements.
177
178 We want the most control dependent block in the shallowest loop nest.
179
180 If the resulting block is in a shallower loop nest, then use it. Else
181 only use the resulting block if it has significantly lower execution
182 frequency than EARLY_BB to avoid gratuitous statement movement. We
183 consider statements with VOPS more desirable to move.
184
185 This pass would obviously benefit from PDO as it utilizes block
186 frequencies. It would also benefit from recomputing frequencies
187 if profile data is not available since frequencies often get out
188 of sync with reality. */
189
190 static basic_block
select_best_block(basic_block early_bb,basic_block late_bb,gimple * stmt)191 select_best_block (basic_block early_bb,
192 basic_block late_bb,
193 gimple *stmt)
194 {
195 basic_block best_bb = late_bb;
196 basic_block temp_bb = late_bb;
197 int threshold;
198
199 while (temp_bb != early_bb)
200 {
201 /* If we've moved into a lower loop nest, then that becomes
202 our best block. */
203 if (bb_loop_depth (temp_bb) < bb_loop_depth (best_bb))
204 best_bb = temp_bb;
205
206 /* Walk up the dominator tree, hopefully we'll find a shallower
207 loop nest. */
208 temp_bb = get_immediate_dominator (CDI_DOMINATORS, temp_bb);
209 }
210
211 /* If we found a shallower loop nest, then we always consider that
212 a win. This will always give us the most control dependent block
213 within that loop nest. */
214 if (bb_loop_depth (best_bb) < bb_loop_depth (early_bb))
215 return best_bb;
216
217 /* Get the sinking threshold. If the statement to be moved has memory
218 operands, then increase the threshold by 7% as those are even more
219 profitable to avoid, clamping at 100%. */
220 threshold = param_sink_frequency_threshold;
221 if (gimple_vuse (stmt) || gimple_vdef (stmt))
222 {
223 threshold += 7;
224 if (threshold > 100)
225 threshold = 100;
226 }
227
228 /* If BEST_BB is at the same nesting level, then require it to have
229 significantly lower execution frequency to avoid gratuitous movement. */
230 if (bb_loop_depth (best_bb) == bb_loop_depth (early_bb)
231 /* If result of comparsion is unknown, prefer EARLY_BB.
232 Thus use !(...>=..) rather than (...<...) */
233 && !(best_bb->count.apply_scale (100, 1)
234 >= early_bb->count.apply_scale (threshold, 1)))
235 return best_bb;
236
237 /* No better block found, so return EARLY_BB, which happens to be the
238 statement's original block. */
239 return early_bb;
240 }
241
242 /* Given a statement (STMT) and the basic block it is currently in (FROMBB),
243 determine the location to sink the statement to, if any.
244 Returns true if there is such location; in that case, TOGSI points to the
245 statement before that STMT should be moved. */
246
247 static bool
statement_sink_location(gimple * stmt,basic_block frombb,gimple_stmt_iterator * togsi,bool * zero_uses_p)248 statement_sink_location (gimple *stmt, basic_block frombb,
249 gimple_stmt_iterator *togsi, bool *zero_uses_p)
250 {
251 gimple *use;
252 use_operand_p one_use = NULL_USE_OPERAND_P;
253 basic_block sinkbb;
254 use_operand_p use_p;
255 def_operand_p def_p;
256 ssa_op_iter iter;
257 imm_use_iterator imm_iter;
258
259 *zero_uses_p = false;
260
261 /* We only can sink assignments and non-looping const/pure calls. */
262 int cf;
263 if (!is_gimple_assign (stmt)
264 && (!is_gimple_call (stmt)
265 || !((cf = gimple_call_flags (stmt)) & (ECF_CONST|ECF_PURE))
266 || (cf & ECF_LOOPING_CONST_OR_PURE)))
267 return false;
268
269 /* We only can sink stmts with a single definition. */
270 def_p = single_ssa_def_operand (stmt, SSA_OP_ALL_DEFS);
271 if (def_p == NULL_DEF_OPERAND_P)
272 return false;
273
274 /* There are a few classes of things we can't or don't move, some because we
275 don't have code to handle it, some because it's not profitable and some
276 because it's not legal.
277
278 We can't sink things that may be global stores, at least not without
279 calculating a lot more information, because we may cause it to no longer
280 be seen by an external routine that needs it depending on where it gets
281 moved to.
282
283 We can't sink statements that end basic blocks without splitting the
284 incoming edge for the sink location to place it there.
285
286 We can't sink statements that have volatile operands.
287
288 We don't want to sink dead code, so anything with 0 immediate uses is not
289 sunk.
290
291 Don't sink BLKmode assignments if current function has any local explicit
292 register variables, as BLKmode assignments may involve memcpy or memset
293 calls or, on some targets, inline expansion thereof that sometimes need
294 to use specific hard registers.
295
296 */
297 if (stmt_ends_bb_p (stmt)
298 || gimple_has_side_effects (stmt)
299 || (cfun->has_local_explicit_reg_vars
300 && TYPE_MODE (TREE_TYPE (gimple_get_lhs (stmt))) == BLKmode))
301 return false;
302
303 /* Return if there are no immediate uses of this stmt. */
304 if (has_zero_uses (DEF_FROM_PTR (def_p)))
305 {
306 *zero_uses_p = true;
307 return false;
308 }
309
310 if (SSA_NAME_OCCURS_IN_ABNORMAL_PHI (DEF_FROM_PTR (def_p)))
311 return false;
312
313 FOR_EACH_SSA_USE_OPERAND (use_p, stmt, iter, SSA_OP_ALL_USES)
314 {
315 tree use = USE_FROM_PTR (use_p);
316 if (SSA_NAME_OCCURS_IN_ABNORMAL_PHI (use))
317 return false;
318 }
319
320 use = NULL;
321
322 /* If stmt is a store the one and only use needs to be the VOP
323 merging PHI node. */
324 if (virtual_operand_p (DEF_FROM_PTR (def_p)))
325 {
326 FOR_EACH_IMM_USE_FAST (use_p, imm_iter, DEF_FROM_PTR (def_p))
327 {
328 gimple *use_stmt = USE_STMT (use_p);
329
330 /* A killing definition is not a use. */
331 if ((gimple_has_lhs (use_stmt)
332 && operand_equal_p (gimple_get_lhs (stmt),
333 gimple_get_lhs (use_stmt), 0))
334 || stmt_kills_ref_p (use_stmt, gimple_get_lhs (stmt)))
335 {
336 /* If use_stmt is or might be a nop assignment then USE_STMT
337 acts as a use as well as definition. */
338 if (stmt != use_stmt
339 && ref_maybe_used_by_stmt_p (use_stmt,
340 gimple_get_lhs (stmt)))
341 return false;
342 continue;
343 }
344
345 if (gimple_code (use_stmt) != GIMPLE_PHI)
346 return false;
347
348 if (use
349 && use != use_stmt)
350 return false;
351
352 use = use_stmt;
353 }
354 if (!use)
355 return false;
356 }
357 /* If all the immediate uses are not in the same place, find the nearest
358 common dominator of all the immediate uses. For PHI nodes, we have to
359 find the nearest common dominator of all of the predecessor blocks, since
360 that is where insertion would have to take place. */
361 else if (gimple_vuse (stmt)
362 || !all_immediate_uses_same_place (def_p))
363 {
364 bool debug_stmts = false;
365 basic_block commondom = nearest_common_dominator_of_uses (def_p,
366 &debug_stmts);
367
368 if (commondom == frombb)
369 return false;
370
371 /* If this is a load then do not sink past any stores.
372 Look for virtual definitions in the path from frombb to the sink
373 location computed from the real uses and if found, adjust
374 that it a common dominator. */
375 if (gimple_vuse (stmt))
376 {
377 /* Do not sink loads from hard registers. */
378 if (gimple_assign_single_p (stmt)
379 && TREE_CODE (gimple_assign_rhs1 (stmt)) == VAR_DECL
380 && DECL_HARD_REGISTER (gimple_assign_rhs1 (stmt)))
381 return false;
382
383 imm_use_iterator imm_iter;
384 use_operand_p use_p;
385 FOR_EACH_IMM_USE_FAST (use_p, imm_iter, gimple_vuse (stmt))
386 {
387 gimple *use_stmt = USE_STMT (use_p);
388 basic_block bb = gimple_bb (use_stmt);
389 /* For PHI nodes the block we know sth about is the incoming block
390 with the use. */
391 if (gimple_code (use_stmt) == GIMPLE_PHI)
392 {
393 /* In case the PHI node post-dominates the current insert
394 location we can disregard it. But make sure it is not
395 dominating it as well as can happen in a CFG cycle. */
396 if (commondom != bb
397 && !dominated_by_p (CDI_DOMINATORS, commondom, bb)
398 && dominated_by_p (CDI_POST_DOMINATORS, commondom, bb)
399 /* If the blocks are possibly within the same irreducible
400 cycle the above check breaks down. */
401 && !((bb->flags & commondom->flags & BB_IRREDUCIBLE_LOOP)
402 && bb->loop_father == commondom->loop_father)
403 && !((commondom->flags & BB_IRREDUCIBLE_LOOP)
404 && flow_loop_nested_p (commondom->loop_father,
405 bb->loop_father))
406 && !((bb->flags & BB_IRREDUCIBLE_LOOP)
407 && flow_loop_nested_p (bb->loop_father,
408 commondom->loop_father)))
409 continue;
410 bb = EDGE_PRED (bb, PHI_ARG_INDEX_FROM_USE (use_p))->src;
411 }
412 else if (!gimple_vdef (use_stmt))
413 continue;
414 /* If the use is not dominated by the path entry it is not on
415 the path. */
416 if (!dominated_by_p (CDI_DOMINATORS, bb, frombb))
417 continue;
418 /* There is no easy way to disregard defs not on the path from
419 frombb to commondom so just consider them all. */
420 commondom = nearest_common_dominator (CDI_DOMINATORS,
421 bb, commondom);
422 if (commondom == frombb)
423 return false;
424 }
425 }
426
427 /* Our common dominator has to be dominated by frombb in order to be a
428 trivially safe place to put this statement, since it has multiple
429 uses. */
430 if (!dominated_by_p (CDI_DOMINATORS, commondom, frombb))
431 return false;
432
433 commondom = select_best_block (frombb, commondom, stmt);
434
435 if (commondom == frombb)
436 return false;
437
438 *togsi = gsi_after_labels (commondom);
439
440 return true;
441 }
442 else
443 {
444 FOR_EACH_IMM_USE_FAST (one_use, imm_iter, DEF_FROM_PTR (def_p))
445 {
446 if (is_gimple_debug (USE_STMT (one_use)))
447 continue;
448 break;
449 }
450 use = USE_STMT (one_use);
451
452 if (gimple_code (use) != GIMPLE_PHI)
453 {
454 sinkbb = select_best_block (frombb, gimple_bb (use), stmt);
455
456 if (sinkbb == frombb)
457 return false;
458
459 if (sinkbb == gimple_bb (use))
460 *togsi = gsi_for_stmt (use);
461 else
462 *togsi = gsi_after_labels (sinkbb);
463
464 return true;
465 }
466 }
467
468 sinkbb = find_bb_for_arg (as_a <gphi *> (use), DEF_FROM_PTR (def_p));
469
470 /* This can happen if there are multiple uses in a PHI. */
471 if (!sinkbb)
472 return false;
473
474 sinkbb = select_best_block (frombb, sinkbb, stmt);
475 if (!sinkbb || sinkbb == frombb)
476 return false;
477
478 /* If the latch block is empty, don't make it non-empty by sinking
479 something into it. */
480 if (sinkbb == frombb->loop_father->latch
481 && empty_block_p (sinkbb))
482 return false;
483
484 *togsi = gsi_after_labels (sinkbb);
485
486 return true;
487 }
488
489 /* Very simplistic code to sink common stores from the predecessor through
490 our virtual PHI. We do this before sinking stmts from BB as it might
491 expose sinking opportunities of the merged stores.
492 Once we have partial dead code elimination through sth like SSU-PRE this
493 should be moved there. */
494
495 static unsigned
sink_common_stores_to_bb(basic_block bb)496 sink_common_stores_to_bb (basic_block bb)
497 {
498 unsigned todo = 0;
499 gphi *phi;
500
501 if (EDGE_COUNT (bb->preds) > 1
502 && (phi = get_virtual_phi (bb)))
503 {
504 /* Repeat until no more common stores are found. */
505 while (1)
506 {
507 gimple *first_store = NULL;
508 auto_vec <tree, 5> vdefs;
509 gimple_stmt_iterator gsi;
510
511 /* Search for common stores defined by all virtual PHI args.
512 ??? Common stores not present in all predecessors could
513 be handled by inserting a forwarder to sink to. Generally
514 this involves deciding which stores to do this for if
515 multiple common stores are present for different sets of
516 predecessors. See PR11832 for an interesting case. */
517 for (unsigned i = 0; i < gimple_phi_num_args (phi); ++i)
518 {
519 tree arg = gimple_phi_arg_def (phi, i);
520 gimple *def = SSA_NAME_DEF_STMT (arg);
521 if (! is_gimple_assign (def)
522 || stmt_can_throw_internal (cfun, def)
523 || (gimple_phi_arg_edge (phi, i)->flags & EDGE_ABNORMAL))
524 {
525 /* ??? We could handle some cascading with the def being
526 another PHI. We'd have to insert multiple PHIs for
527 the rhs then though (if they are not all equal). */
528 first_store = NULL;
529 break;
530 }
531 /* ??? Do not try to do anything fancy with aliasing, thus
532 do not sink across non-aliased loads (or even stores,
533 so different store order will make the sinking fail). */
534 bool all_uses_on_phi = true;
535 imm_use_iterator iter;
536 use_operand_p use_p;
537 FOR_EACH_IMM_USE_FAST (use_p, iter, arg)
538 if (USE_STMT (use_p) != phi)
539 {
540 all_uses_on_phi = false;
541 break;
542 }
543 if (! all_uses_on_phi)
544 {
545 first_store = NULL;
546 break;
547 }
548 /* Check all stores are to the same LHS. */
549 if (! first_store)
550 first_store = def;
551 /* ??? We could handle differing SSA uses in the LHS by inserting
552 PHIs for them. */
553 else if (! operand_equal_p (gimple_assign_lhs (first_store),
554 gimple_assign_lhs (def), 0)
555 || (gimple_clobber_p (first_store)
556 != gimple_clobber_p (def)))
557 {
558 first_store = NULL;
559 break;
560 }
561 vdefs.safe_push (arg);
562 }
563 if (! first_store)
564 break;
565
566 /* Check if we need a PHI node to merge the stored values. */
567 bool allsame = true;
568 if (!gimple_clobber_p (first_store))
569 for (unsigned i = 1; i < vdefs.length (); ++i)
570 {
571 gimple *def = SSA_NAME_DEF_STMT (vdefs[i]);
572 if (! operand_equal_p (gimple_assign_rhs1 (first_store),
573 gimple_assign_rhs1 (def), 0))
574 {
575 allsame = false;
576 break;
577 }
578 }
579
580 /* We cannot handle aggregate values if we need to merge them. */
581 tree type = TREE_TYPE (gimple_assign_lhs (first_store));
582 if (! allsame
583 && ! is_gimple_reg_type (type))
584 break;
585
586 if (dump_enabled_p ())
587 {
588 dump_printf_loc (MSG_OPTIMIZED_LOCATIONS,
589 first_store,
590 "sinking common stores %sto ",
591 allsame ? "with same value " : "");
592 dump_generic_expr (MSG_OPTIMIZED_LOCATIONS, TDF_SLIM,
593 gimple_assign_lhs (first_store));
594 dump_printf (MSG_OPTIMIZED_LOCATIONS, "\n");
595 }
596
597 /* Insert a PHI to merge differing stored values if necessary.
598 Note that in general inserting PHIs isn't a very good idea as
599 it makes the job of coalescing and register allocation harder.
600 Even common SSA uses on the rhs/lhs might extend their lifetime
601 across multiple edges by this code motion which makes
602 register allocation harder. */
603 tree from;
604 if (! allsame)
605 {
606 from = make_ssa_name (type);
607 gphi *newphi = create_phi_node (from, bb);
608 for (unsigned i = 0; i < vdefs.length (); ++i)
609 {
610 gimple *def = SSA_NAME_DEF_STMT (vdefs[i]);
611 add_phi_arg (newphi, gimple_assign_rhs1 (def),
612 EDGE_PRED (bb, i), UNKNOWN_LOCATION);
613 }
614 }
615 else
616 from = gimple_assign_rhs1 (first_store);
617
618 /* Remove all stores. */
619 for (unsigned i = 0; i < vdefs.length (); ++i)
620 TREE_VISITED (vdefs[i]) = 1;
621 for (unsigned i = 0; i < vdefs.length (); ++i)
622 /* If we have more than one use of a VDEF on the PHI make sure
623 we remove the defining stmt only once. */
624 if (TREE_VISITED (vdefs[i]))
625 {
626 TREE_VISITED (vdefs[i]) = 0;
627 gimple *def = SSA_NAME_DEF_STMT (vdefs[i]);
628 gsi = gsi_for_stmt (def);
629 unlink_stmt_vdef (def);
630 gsi_remove (&gsi, true);
631 release_defs (def);
632 }
633
634 /* Insert the first store at the beginning of the merge BB. */
635 gimple_set_vdef (first_store, gimple_phi_result (phi));
636 SSA_NAME_DEF_STMT (gimple_vdef (first_store)) = first_store;
637 gimple_phi_set_result (phi, make_ssa_name (gimple_vop (cfun)));
638 gimple_set_vuse (first_store, gimple_phi_result (phi));
639 gimple_assign_set_rhs1 (first_store, from);
640 /* ??? Should we reset first_stores location? */
641 gsi = gsi_after_labels (bb);
642 gsi_insert_before (&gsi, first_store, GSI_SAME_STMT);
643 sink_stats.commoned++;
644
645 todo |= TODO_cleanup_cfg;
646 }
647
648 /* We could now have empty predecessors that we could remove,
649 forming a proper CFG for further sinking. Note that even
650 CFG cleanup doesn't do this fully at the moment and it
651 doesn't preserve post-dominators in the process either.
652 The mergephi pass might do it though. gcc.dg/tree-ssa/ssa-sink-13.c
653 shows this nicely if you disable tail merging or (same effect)
654 make the stored values unequal. */
655 }
656
657 return todo;
658 }
659
660 /* Perform code sinking on BB */
661
662 static unsigned
sink_code_in_bb(basic_block bb)663 sink_code_in_bb (basic_block bb)
664 {
665 basic_block son;
666 gimple_stmt_iterator gsi;
667 edge_iterator ei;
668 edge e;
669 bool last = true;
670 unsigned todo = 0;
671
672 /* Sink common stores from the predecessor through our virtual PHI. */
673 todo |= sink_common_stores_to_bb (bb);
674
675 /* If this block doesn't dominate anything, there can't be any place to sink
676 the statements to. */
677 if (first_dom_son (CDI_DOMINATORS, bb) == NULL)
678 goto earlyout;
679
680 /* We can't move things across abnormal edges, so don't try. */
681 FOR_EACH_EDGE (e, ei, bb->succs)
682 if (e->flags & EDGE_ABNORMAL)
683 goto earlyout;
684
685 for (gsi = gsi_last_bb (bb); !gsi_end_p (gsi);)
686 {
687 gimple *stmt = gsi_stmt (gsi);
688 gimple_stmt_iterator togsi;
689 bool zero_uses_p;
690
691 if (!statement_sink_location (stmt, bb, &togsi, &zero_uses_p))
692 {
693 gimple_stmt_iterator saved = gsi;
694 if (!gsi_end_p (gsi))
695 gsi_prev (&gsi);
696 /* If we face a dead stmt remove it as it possibly blocks
697 sinking of uses. */
698 if (zero_uses_p
699 && ! gimple_vdef (stmt))
700 {
701 gsi_remove (&saved, true);
702 release_defs (stmt);
703 }
704 else
705 last = false;
706 continue;
707 }
708 if (dump_file)
709 {
710 fprintf (dump_file, "Sinking ");
711 print_gimple_stmt (dump_file, stmt, 0, TDF_VOPS);
712 fprintf (dump_file, " from bb %d to bb %d\n",
713 bb->index, (gsi_bb (togsi))->index);
714 }
715
716 /* Update virtual operands of statements in the path we
717 do not sink to. */
718 if (gimple_vdef (stmt))
719 {
720 imm_use_iterator iter;
721 use_operand_p use_p;
722 gimple *vuse_stmt;
723
724 FOR_EACH_IMM_USE_STMT (vuse_stmt, iter, gimple_vdef (stmt))
725 if (gimple_code (vuse_stmt) != GIMPLE_PHI)
726 FOR_EACH_IMM_USE_ON_STMT (use_p, iter)
727 SET_USE (use_p, gimple_vuse (stmt));
728 }
729
730 /* If this is the end of the basic block, we need to insert at the end
731 of the basic block. */
732 if (gsi_end_p (togsi))
733 gsi_move_to_bb_end (&gsi, gsi_bb (togsi));
734 else
735 gsi_move_before (&gsi, &togsi);
736
737 sink_stats.sunk++;
738
739 /* If we've just removed the last statement of the BB, the
740 gsi_end_p() test below would fail, but gsi_prev() would have
741 succeeded, and we want it to succeed. So we keep track of
742 whether we're at the last statement and pick up the new last
743 statement. */
744 if (last)
745 {
746 gsi = gsi_last_bb (bb);
747 continue;
748 }
749
750 last = false;
751 if (!gsi_end_p (gsi))
752 gsi_prev (&gsi);
753
754 }
755 earlyout:
756 for (son = first_dom_son (CDI_POST_DOMINATORS, bb);
757 son;
758 son = next_dom_son (CDI_POST_DOMINATORS, son))
759 {
760 todo |= sink_code_in_bb (son);
761 }
762
763 return todo;
764 }
765
766 /* Perform code sinking.
767 This moves code down the flowgraph when we know it would be
768 profitable to do so, or it wouldn't increase the number of
769 executions of the statement.
770
771 IE given
772
773 a_1 = b + c;
774 if (<something>)
775 {
776 }
777 else
778 {
779 foo (&b, &c);
780 a_5 = b + c;
781 }
782 a_6 = PHI (a_5, a_1);
783 USE a_6.
784
785 we'll transform this into:
786
787 if (<something>)
788 {
789 a_1 = b + c;
790 }
791 else
792 {
793 foo (&b, &c);
794 a_5 = b + c;
795 }
796 a_6 = PHI (a_5, a_1);
797 USE a_6.
798
799 Note that this reduces the number of computations of a = b + c to 1
800 when we take the else edge, instead of 2.
801 */
802 namespace {
803
804 const pass_data pass_data_sink_code =
805 {
806 GIMPLE_PASS, /* type */
807 "sink", /* name */
808 OPTGROUP_NONE, /* optinfo_flags */
809 TV_TREE_SINK, /* tv_id */
810 /* PROP_no_crit_edges is ensured by running split_edges_for_insertion in
811 pass_data_sink_code::execute (). */
812 ( PROP_cfg | PROP_ssa ), /* properties_required */
813 0, /* properties_provided */
814 0, /* properties_destroyed */
815 0, /* todo_flags_start */
816 TODO_update_ssa, /* todo_flags_finish */
817 };
818
819 class pass_sink_code : public gimple_opt_pass
820 {
821 public:
pass_sink_code(gcc::context * ctxt)822 pass_sink_code (gcc::context *ctxt)
823 : gimple_opt_pass (pass_data_sink_code, ctxt)
824 {}
825
826 /* opt_pass methods: */
gate(function *)827 virtual bool gate (function *) { return flag_tree_sink != 0; }
828 virtual unsigned int execute (function *);
829
830 }; // class pass_sink_code
831
832 unsigned int
execute(function * fun)833 pass_sink_code::execute (function *fun)
834 {
835 loop_optimizer_init (LOOPS_NORMAL);
836 split_edges_for_insertion ();
837 connect_infinite_loops_to_exit ();
838 memset (&sink_stats, 0, sizeof (sink_stats));
839 calculate_dominance_info (CDI_DOMINATORS);
840 calculate_dominance_info (CDI_POST_DOMINATORS);
841 unsigned todo = sink_code_in_bb (EXIT_BLOCK_PTR_FOR_FN (fun));
842 statistics_counter_event (fun, "Sunk statements", sink_stats.sunk);
843 statistics_counter_event (fun, "Commoned stores", sink_stats.commoned);
844 free_dominance_info (CDI_POST_DOMINATORS);
845 remove_fake_exit_edges ();
846 loop_optimizer_finalize ();
847
848 return todo;
849 }
850
851 } // anon namespace
852
853 gimple_opt_pass *
make_pass_sink_code(gcc::context * ctxt)854 make_pass_sink_code (gcc::context *ctxt)
855 {
856 return new pass_sink_code (ctxt);
857 }
858