Daily bump.
[official-gcc.git] / gcc / tree-parloops.c
blob3a788ccf1b74083b07e0e6347870bde1ccddb444
1 /* Loop autoparallelization.
2 Copyright (C) 2006-2018 Free Software Foundation, Inc.
3 Contributed by Sebastian Pop <pop@cri.ensmp.fr>
4 Zdenek Dvorak <dvorakz@suse.cz> and Razya Ladelsky <razya@il.ibm.com>.
6 This file is part of GCC.
8 GCC is free software; you can redistribute it and/or modify it under
9 the terms of the GNU General Public License as published by the Free
10 Software Foundation; either version 3, or (at your option) any later
11 version.
13 GCC is distributed in the hope that it will be useful, but WITHOUT ANY
14 WARRANTY; without even the implied warranty of MERCHANTABILITY or
15 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
16 for more details.
18 You should have received a copy of the GNU General Public License
19 along with GCC; see the file COPYING3. If not see
20 <http://www.gnu.org/licenses/>. */
22 #include "config.h"
23 #include "system.h"
24 #include "coretypes.h"
25 #include "backend.h"
26 #include "tree.h"
27 #include "gimple.h"
28 #include "cfghooks.h"
29 #include "tree-pass.h"
30 #include "ssa.h"
31 #include "cgraph.h"
32 #include "gimple-pretty-print.h"
33 #include "fold-const.h"
34 #include "gimplify.h"
35 #include "gimple-iterator.h"
36 #include "gimplify-me.h"
37 #include "gimple-walk.h"
38 #include "stor-layout.h"
39 #include "tree-nested.h"
40 #include "tree-cfg.h"
41 #include "tree-ssa-loop-ivopts.h"
42 #include "tree-ssa-loop-manip.h"
43 #include "tree-ssa-loop-niter.h"
44 #include "tree-ssa-loop.h"
45 #include "tree-into-ssa.h"
46 #include "cfgloop.h"
47 #include "tree-scalar-evolution.h"
48 #include "langhooks.h"
49 #include "tree-vectorizer.h"
50 #include "tree-hasher.h"
51 #include "tree-parloops.h"
52 #include "omp-general.h"
53 #include "omp-low.h"
54 #include "tree-ssa.h"
55 #include "params.h"
56 #include "params-enum.h"
57 #include "tree-ssa-alias.h"
58 #include "tree-eh.h"
59 #include "gomp-constants.h"
60 #include "tree-dfa.h"
61 #include "stringpool.h"
62 #include "attribs.h"
64 /* This pass tries to distribute iterations of loops into several threads.
65 The implementation is straightforward -- for each loop we test whether its
66 iterations are independent, and if it is the case (and some additional
67 conditions regarding profitability and correctness are satisfied), we
68 add GIMPLE_OMP_PARALLEL and GIMPLE_OMP_FOR codes and let omp expansion
69 machinery do its job.
71 The most of the complexity is in bringing the code into shape expected
72 by the omp expanders:
73 -- for GIMPLE_OMP_FOR, ensuring that the loop has only one induction
74 variable and that the exit test is at the start of the loop body
75 -- for GIMPLE_OMP_PARALLEL, replacing the references to local addressable
76 variables by accesses through pointers, and breaking up ssa chains
77 by storing the values incoming to the parallelized loop to a structure
78 passed to the new function as an argument (something similar is done
79 in omp gimplification, unfortunately only a small part of the code
80 can be shared).
82 TODO:
83 -- if there are several parallelizable loops in a function, it may be
84 possible to generate the threads just once (using synchronization to
85 ensure that cross-loop dependences are obeyed).
86 -- handling of common reduction patterns for outer loops.
88 More info can also be found at http://gcc.gnu.org/wiki/AutoParInGCC */
90 Reduction handling:
91 currently we use vect_force_simple_reduction() to detect reduction patterns.
92 The code transformation will be introduced by an example.
95 parloop
97 int sum=1;
99 for (i = 0; i < N; i++)
101 x[i] = i + 3;
102 sum+=x[i];
106 gimple-like code:
107 header_bb:
109 # sum_29 = PHI <sum_11(5), 1(3)>
110 # i_28 = PHI <i_12(5), 0(3)>
111 D.1795_8 = i_28 + 3;
112 x[i_28] = D.1795_8;
113 sum_11 = D.1795_8 + sum_29;
114 i_12 = i_28 + 1;
115 if (N_6(D) > i_12)
116 goto header_bb;
119 exit_bb:
121 # sum_21 = PHI <sum_11(4)>
122 printf (&"%d"[0], sum_21);
125 after reduction transformation (only relevant parts):
127 parloop
130 ....
133 # Storing the initial value given by the user. #
135 .paral_data_store.32.sum.27 = 1;
137 #pragma omp parallel num_threads(4)
139 #pragma omp for schedule(static)
141 # The neutral element corresponding to the particular
142 reduction's operation, e.g. 0 for PLUS_EXPR,
143 1 for MULT_EXPR, etc. replaces the user's initial value. #
145 # sum.27_29 = PHI <sum.27_11, 0>
147 sum.27_11 = D.1827_8 + sum.27_29;
149 GIMPLE_OMP_CONTINUE
151 # Adding this reduction phi is done at create_phi_for_local_result() #
152 # sum.27_56 = PHI <sum.27_11, 0>
153 GIMPLE_OMP_RETURN
155 # Creating the atomic operation is done at
156 create_call_for_reduction_1() #
158 #pragma omp atomic_load
159 D.1839_59 = *&.paral_data_load.33_51->reduction.23;
160 D.1840_60 = sum.27_56 + D.1839_59;
161 #pragma omp atomic_store (D.1840_60);
163 GIMPLE_OMP_RETURN
165 # collecting the result after the join of the threads is done at
166 create_loads_for_reductions().
167 The value computed by the threads is loaded from the
168 shared struct. #
171 .paral_data_load.33_52 = &.paral_data_store.32;
172 sum_37 = .paral_data_load.33_52->sum.27;
173 sum_43 = D.1795_41 + sum_37;
175 exit bb:
176 # sum_21 = PHI <sum_43, sum_26>
177 printf (&"%d"[0], sum_21);
185 /* Minimal number of iterations of a loop that should be executed in each
186 thread. */
187 #define MIN_PER_THREAD PARAM_VALUE (PARAM_PARLOOPS_MIN_PER_THREAD)
189 /* Element of the hashtable, representing a
190 reduction in the current loop. */
191 struct reduction_info
193 gimple *reduc_stmt; /* reduction statement. */
194 gimple *reduc_phi; /* The phi node defining the reduction. */
195 enum tree_code reduction_code;/* code for the reduction operation. */
196 unsigned reduc_version; /* SSA_NAME_VERSION of original reduc_phi
197 result. */
198 gphi *keep_res; /* The PHI_RESULT of this phi is the resulting value
199 of the reduction variable when existing the loop. */
200 tree initial_value; /* The initial value of the reduction var before entering the loop. */
201 tree field; /* the name of the field in the parloop data structure intended for reduction. */
202 tree reduc_addr; /* The address of the reduction variable for
203 openacc reductions. */
204 tree init; /* reduction initialization value. */
205 gphi *new_phi; /* (helper field) Newly created phi node whose result
206 will be passed to the atomic operation. Represents
207 the local result each thread computed for the reduction
208 operation. */
211 /* Reduction info hashtable helpers. */
213 struct reduction_hasher : free_ptr_hash <reduction_info>
215 static inline hashval_t hash (const reduction_info *);
216 static inline bool equal (const reduction_info *, const reduction_info *);
219 /* Equality and hash functions for hashtab code. */
221 inline bool
222 reduction_hasher::equal (const reduction_info *a, const reduction_info *b)
224 return (a->reduc_phi == b->reduc_phi);
227 inline hashval_t
228 reduction_hasher::hash (const reduction_info *a)
230 return a->reduc_version;
233 typedef hash_table<reduction_hasher> reduction_info_table_type;
236 static struct reduction_info *
237 reduction_phi (reduction_info_table_type *reduction_list, gimple *phi)
239 struct reduction_info tmpred, *red;
241 if (reduction_list->elements () == 0 || phi == NULL)
242 return NULL;
244 if (gimple_uid (phi) == (unsigned int)-1
245 || gimple_uid (phi) == 0)
246 return NULL;
248 tmpred.reduc_phi = phi;
249 tmpred.reduc_version = gimple_uid (phi);
250 red = reduction_list->find (&tmpred);
251 gcc_assert (red == NULL || red->reduc_phi == phi);
253 return red;
256 /* Element of hashtable of names to copy. */
258 struct name_to_copy_elt
260 unsigned version; /* The version of the name to copy. */
261 tree new_name; /* The new name used in the copy. */
262 tree field; /* The field of the structure used to pass the
263 value. */
266 /* Name copies hashtable helpers. */
268 struct name_to_copy_hasher : free_ptr_hash <name_to_copy_elt>
270 static inline hashval_t hash (const name_to_copy_elt *);
271 static inline bool equal (const name_to_copy_elt *, const name_to_copy_elt *);
274 /* Equality and hash functions for hashtab code. */
276 inline bool
277 name_to_copy_hasher::equal (const name_to_copy_elt *a, const name_to_copy_elt *b)
279 return a->version == b->version;
282 inline hashval_t
283 name_to_copy_hasher::hash (const name_to_copy_elt *a)
285 return (hashval_t) a->version;
288 typedef hash_table<name_to_copy_hasher> name_to_copy_table_type;
290 /* A transformation matrix, which is a self-contained ROWSIZE x COLSIZE
291 matrix. Rather than use floats, we simply keep a single DENOMINATOR that
292 represents the denominator for every element in the matrix. */
293 typedef struct lambda_trans_matrix_s
295 lambda_matrix matrix;
296 int rowsize;
297 int colsize;
298 int denominator;
299 } *lambda_trans_matrix;
300 #define LTM_MATRIX(T) ((T)->matrix)
301 #define LTM_ROWSIZE(T) ((T)->rowsize)
302 #define LTM_COLSIZE(T) ((T)->colsize)
303 #define LTM_DENOMINATOR(T) ((T)->denominator)
305 /* Allocate a new transformation matrix. */
307 static lambda_trans_matrix
308 lambda_trans_matrix_new (int colsize, int rowsize,
309 struct obstack * lambda_obstack)
311 lambda_trans_matrix ret;
313 ret = (lambda_trans_matrix)
314 obstack_alloc (lambda_obstack, sizeof (struct lambda_trans_matrix_s));
315 LTM_MATRIX (ret) = lambda_matrix_new (rowsize, colsize, lambda_obstack);
316 LTM_ROWSIZE (ret) = rowsize;
317 LTM_COLSIZE (ret) = colsize;
318 LTM_DENOMINATOR (ret) = 1;
319 return ret;
322 /* Multiply a vector VEC by a matrix MAT.
323 MAT is an M*N matrix, and VEC is a vector with length N. The result
324 is stored in DEST which must be a vector of length M. */
326 static void
327 lambda_matrix_vector_mult (lambda_matrix matrix, int m, int n,
328 lambda_vector vec, lambda_vector dest)
330 int i, j;
332 lambda_vector_clear (dest, m);
333 for (i = 0; i < m; i++)
334 for (j = 0; j < n; j++)
335 dest[i] += matrix[i][j] * vec[j];
338 /* Return true if TRANS is a legal transformation matrix that respects
339 the dependence vectors in DISTS and DIRS. The conservative answer
340 is false.
342 "Wolfe proves that a unimodular transformation represented by the
343 matrix T is legal when applied to a loop nest with a set of
344 lexicographically non-negative distance vectors RDG if and only if
345 for each vector d in RDG, (T.d >= 0) is lexicographically positive.
346 i.e.: if and only if it transforms the lexicographically positive
347 distance vectors to lexicographically positive vectors. Note that
348 a unimodular matrix must transform the zero vector (and only it) to
349 the zero vector." S.Muchnick. */
351 static bool
352 lambda_transform_legal_p (lambda_trans_matrix trans,
353 int nb_loops,
354 vec<ddr_p> dependence_relations)
356 unsigned int i, j;
357 lambda_vector distres;
358 struct data_dependence_relation *ddr;
360 gcc_assert (LTM_COLSIZE (trans) == nb_loops
361 && LTM_ROWSIZE (trans) == nb_loops);
363 /* When there are no dependences, the transformation is correct. */
364 if (dependence_relations.length () == 0)
365 return true;
367 ddr = dependence_relations[0];
368 if (ddr == NULL)
369 return true;
371 /* When there is an unknown relation in the dependence_relations, we
372 know that it is no worth looking at this loop nest: give up. */
373 if (DDR_ARE_DEPENDENT (ddr) == chrec_dont_know)
374 return false;
376 distres = lambda_vector_new (nb_loops);
378 /* For each distance vector in the dependence graph. */
379 FOR_EACH_VEC_ELT (dependence_relations, i, ddr)
381 /* Don't care about relations for which we know that there is no
382 dependence, nor about read-read (aka. output-dependences):
383 these data accesses can happen in any order. */
384 if (DDR_ARE_DEPENDENT (ddr) == chrec_known
385 || (DR_IS_READ (DDR_A (ddr)) && DR_IS_READ (DDR_B (ddr))))
386 continue;
388 /* Conservatively answer: "this transformation is not valid". */
389 if (DDR_ARE_DEPENDENT (ddr) == chrec_dont_know)
390 return false;
392 /* If the dependence could not be captured by a distance vector,
393 conservatively answer that the transform is not valid. */
394 if (DDR_NUM_DIST_VECTS (ddr) == 0)
395 return false;
397 /* Compute trans.dist_vect */
398 for (j = 0; j < DDR_NUM_DIST_VECTS (ddr); j++)
400 lambda_matrix_vector_mult (LTM_MATRIX (trans), nb_loops, nb_loops,
401 DDR_DIST_VECT (ddr, j), distres);
403 if (!lambda_vector_lexico_pos (distres, nb_loops))
404 return false;
407 return true;
410 /* Data dependency analysis. Returns true if the iterations of LOOP
411 are independent on each other (that is, if we can execute them
412 in parallel). */
414 static bool
415 loop_parallel_p (struct loop *loop, struct obstack * parloop_obstack)
417 vec<ddr_p> dependence_relations;
418 vec<data_reference_p> datarefs;
419 lambda_trans_matrix trans;
420 bool ret = false;
422 if (dump_file && (dump_flags & TDF_DETAILS))
424 fprintf (dump_file, "Considering loop %d\n", loop->num);
425 if (!loop->inner)
426 fprintf (dump_file, "loop is innermost\n");
427 else
428 fprintf (dump_file, "loop NOT innermost\n");
431 /* Check for problems with dependences. If the loop can be reversed,
432 the iterations are independent. */
433 auto_vec<loop_p, 3> loop_nest;
434 datarefs.create (10);
435 dependence_relations.create (100);
436 if (! compute_data_dependences_for_loop (loop, true, &loop_nest, &datarefs,
437 &dependence_relations))
439 if (dump_file && (dump_flags & TDF_DETAILS))
440 fprintf (dump_file, " FAILED: cannot analyze data dependencies\n");
441 ret = false;
442 goto end;
444 if (dump_file && (dump_flags & TDF_DETAILS))
445 dump_data_dependence_relations (dump_file, dependence_relations);
447 trans = lambda_trans_matrix_new (1, 1, parloop_obstack);
448 LTM_MATRIX (trans)[0][0] = -1;
450 if (lambda_transform_legal_p (trans, 1, dependence_relations))
452 ret = true;
453 if (dump_file && (dump_flags & TDF_DETAILS))
454 fprintf (dump_file, " SUCCESS: may be parallelized\n");
456 else if (dump_file && (dump_flags & TDF_DETAILS))
457 fprintf (dump_file,
458 " FAILED: data dependencies exist across iterations\n");
460 end:
461 free_dependence_relations (dependence_relations);
462 free_data_refs (datarefs);
464 return ret;
467 /* Return true when LOOP contains basic blocks marked with the
468 BB_IRREDUCIBLE_LOOP flag. */
470 static inline bool
471 loop_has_blocks_with_irreducible_flag (struct loop *loop)
473 unsigned i;
474 basic_block *bbs = get_loop_body_in_dom_order (loop);
475 bool res = true;
477 for (i = 0; i < loop->num_nodes; i++)
478 if (bbs[i]->flags & BB_IRREDUCIBLE_LOOP)
479 goto end;
481 res = false;
482 end:
483 free (bbs);
484 return res;
487 /* Assigns the address of OBJ in TYPE to an ssa name, and returns this name.
488 The assignment statement is placed on edge ENTRY. DECL_ADDRESS maps decls
489 to their addresses that can be reused. The address of OBJ is known to
490 be invariant in the whole function. Other needed statements are placed
491 right before GSI. */
493 static tree
494 take_address_of (tree obj, tree type, edge entry,
495 int_tree_htab_type *decl_address, gimple_stmt_iterator *gsi)
497 int uid;
498 tree *var_p, name, addr;
499 gassign *stmt;
500 gimple_seq stmts;
502 /* Since the address of OBJ is invariant, the trees may be shared.
503 Avoid rewriting unrelated parts of the code. */
504 obj = unshare_expr (obj);
505 for (var_p = &obj;
506 handled_component_p (*var_p);
507 var_p = &TREE_OPERAND (*var_p, 0))
508 continue;
510 /* Canonicalize the access to base on a MEM_REF. */
511 if (DECL_P (*var_p))
512 *var_p = build_simple_mem_ref (build_fold_addr_expr (*var_p));
514 /* Assign a canonical SSA name to the address of the base decl used
515 in the address and share it for all accesses and addresses based
516 on it. */
517 uid = DECL_UID (TREE_OPERAND (TREE_OPERAND (*var_p, 0), 0));
518 int_tree_map elt;
519 elt.uid = uid;
520 int_tree_map *slot = decl_address->find_slot (elt, INSERT);
521 if (!slot->to)
523 if (gsi == NULL)
524 return NULL;
525 addr = TREE_OPERAND (*var_p, 0);
526 const char *obj_name
527 = get_name (TREE_OPERAND (TREE_OPERAND (*var_p, 0), 0));
528 if (obj_name)
529 name = make_temp_ssa_name (TREE_TYPE (addr), NULL, obj_name);
530 else
531 name = make_ssa_name (TREE_TYPE (addr));
532 stmt = gimple_build_assign (name, addr);
533 gsi_insert_on_edge_immediate (entry, stmt);
535 slot->uid = uid;
536 slot->to = name;
538 else
539 name = slot->to;
541 /* Express the address in terms of the canonical SSA name. */
542 TREE_OPERAND (*var_p, 0) = name;
543 if (gsi == NULL)
544 return build_fold_addr_expr_with_type (obj, type);
546 name = force_gimple_operand (build_addr (obj),
547 &stmts, true, NULL_TREE);
548 if (!gimple_seq_empty_p (stmts))
549 gsi_insert_seq_before (gsi, stmts, GSI_SAME_STMT);
551 if (!useless_type_conversion_p (type, TREE_TYPE (name)))
553 name = force_gimple_operand (fold_convert (type, name), &stmts, true,
554 NULL_TREE);
555 if (!gimple_seq_empty_p (stmts))
556 gsi_insert_seq_before (gsi, stmts, GSI_SAME_STMT);
559 return name;
562 static tree
563 reduc_stmt_res (gimple *stmt)
565 return (gimple_code (stmt) == GIMPLE_PHI
566 ? gimple_phi_result (stmt)
567 : gimple_assign_lhs (stmt));
570 /* Callback for htab_traverse. Create the initialization statement
571 for reduction described in SLOT, and place it at the preheader of
572 the loop described in DATA. */
575 initialize_reductions (reduction_info **slot, struct loop *loop)
577 tree init;
578 tree type, arg;
579 edge e;
581 struct reduction_info *const reduc = *slot;
583 /* Create initialization in preheader:
584 reduction_variable = initialization value of reduction. */
586 /* In the phi node at the header, replace the argument coming
587 from the preheader with the reduction initialization value. */
589 /* Initialize the reduction. */
590 type = TREE_TYPE (PHI_RESULT (reduc->reduc_phi));
591 init = omp_reduction_init_op (gimple_location (reduc->reduc_stmt),
592 reduc->reduction_code, type);
593 reduc->init = init;
595 /* Replace the argument representing the initialization value
596 with the initialization value for the reduction (neutral
597 element for the particular operation, e.g. 0 for PLUS_EXPR,
598 1 for MULT_EXPR, etc).
599 Keep the old value in a new variable "reduction_initial",
600 that will be taken in consideration after the parallel
601 computing is done. */
603 e = loop_preheader_edge (loop);
604 arg = PHI_ARG_DEF_FROM_EDGE (reduc->reduc_phi, e);
605 /* Create new variable to hold the initial value. */
607 SET_USE (PHI_ARG_DEF_PTR_FROM_EDGE
608 (reduc->reduc_phi, loop_preheader_edge (loop)), init);
609 reduc->initial_value = arg;
610 return 1;
613 struct elv_data
615 struct walk_stmt_info info;
616 edge entry;
617 int_tree_htab_type *decl_address;
618 gimple_stmt_iterator *gsi;
619 bool changed;
620 bool reset;
623 /* Eliminates references to local variables in *TP out of the single
624 entry single exit region starting at DTA->ENTRY.
625 DECL_ADDRESS contains addresses of the references that had their
626 address taken already. If the expression is changed, CHANGED is
627 set to true. Callback for walk_tree. */
629 static tree
630 eliminate_local_variables_1 (tree *tp, int *walk_subtrees, void *data)
632 struct elv_data *const dta = (struct elv_data *) data;
633 tree t = *tp, var, addr, addr_type, type, obj;
635 if (DECL_P (t))
637 *walk_subtrees = 0;
639 if (!SSA_VAR_P (t) || DECL_EXTERNAL (t))
640 return NULL_TREE;
642 type = TREE_TYPE (t);
643 addr_type = build_pointer_type (type);
644 addr = take_address_of (t, addr_type, dta->entry, dta->decl_address,
645 dta->gsi);
646 if (dta->gsi == NULL && addr == NULL_TREE)
648 dta->reset = true;
649 return NULL_TREE;
652 *tp = build_simple_mem_ref (addr);
654 dta->changed = true;
655 return NULL_TREE;
658 if (TREE_CODE (t) == ADDR_EXPR)
660 /* ADDR_EXPR may appear in two contexts:
661 -- as a gimple operand, when the address taken is a function invariant
662 -- as gimple rhs, when the resulting address in not a function
663 invariant
664 We do not need to do anything special in the latter case (the base of
665 the memory reference whose address is taken may be replaced in the
666 DECL_P case). The former case is more complicated, as we need to
667 ensure that the new address is still a gimple operand. Thus, it
668 is not sufficient to replace just the base of the memory reference --
669 we need to move the whole computation of the address out of the
670 loop. */
671 if (!is_gimple_val (t))
672 return NULL_TREE;
674 *walk_subtrees = 0;
675 obj = TREE_OPERAND (t, 0);
676 var = get_base_address (obj);
677 if (!var || !SSA_VAR_P (var) || DECL_EXTERNAL (var))
678 return NULL_TREE;
680 addr_type = TREE_TYPE (t);
681 addr = take_address_of (obj, addr_type, dta->entry, dta->decl_address,
682 dta->gsi);
683 if (dta->gsi == NULL && addr == NULL_TREE)
685 dta->reset = true;
686 return NULL_TREE;
688 *tp = addr;
690 dta->changed = true;
691 return NULL_TREE;
694 if (!EXPR_P (t))
695 *walk_subtrees = 0;
697 return NULL_TREE;
700 /* Moves the references to local variables in STMT at *GSI out of the single
701 entry single exit region starting at ENTRY. DECL_ADDRESS contains
702 addresses of the references that had their address taken
703 already. */
705 static void
706 eliminate_local_variables_stmt (edge entry, gimple_stmt_iterator *gsi,
707 int_tree_htab_type *decl_address)
709 struct elv_data dta;
710 gimple *stmt = gsi_stmt (*gsi);
712 memset (&dta.info, '\0', sizeof (dta.info));
713 dta.entry = entry;
714 dta.decl_address = decl_address;
715 dta.changed = false;
716 dta.reset = false;
718 if (gimple_debug_bind_p (stmt))
720 dta.gsi = NULL;
721 walk_tree (gimple_debug_bind_get_value_ptr (stmt),
722 eliminate_local_variables_1, &dta.info, NULL);
723 if (dta.reset)
725 gimple_debug_bind_reset_value (stmt);
726 dta.changed = true;
729 else if (gimple_clobber_p (stmt))
731 unlink_stmt_vdef (stmt);
732 stmt = gimple_build_nop ();
733 gsi_replace (gsi, stmt, false);
734 dta.changed = true;
736 else
738 dta.gsi = gsi;
739 walk_gimple_op (stmt, eliminate_local_variables_1, &dta.info);
742 if (dta.changed)
743 update_stmt (stmt);
746 /* Eliminates the references to local variables from the single entry
747 single exit region between the ENTRY and EXIT edges.
749 This includes:
750 1) Taking address of a local variable -- these are moved out of the
751 region (and temporary variable is created to hold the address if
752 necessary).
754 2) Dereferencing a local variable -- these are replaced with indirect
755 references. */
757 static void
758 eliminate_local_variables (edge entry, edge exit)
760 basic_block bb;
761 auto_vec<basic_block, 3> body;
762 unsigned i;
763 gimple_stmt_iterator gsi;
764 bool has_debug_stmt = false;
765 int_tree_htab_type decl_address (10);
766 basic_block entry_bb = entry->src;
767 basic_block exit_bb = exit->dest;
769 gather_blocks_in_sese_region (entry_bb, exit_bb, &body);
771 FOR_EACH_VEC_ELT (body, i, bb)
772 if (bb != entry_bb && bb != exit_bb)
774 for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi))
775 if (is_gimple_debug (gsi_stmt (gsi)))
777 if (gimple_debug_bind_p (gsi_stmt (gsi)))
778 has_debug_stmt = true;
780 else
781 eliminate_local_variables_stmt (entry, &gsi, &decl_address);
784 if (has_debug_stmt)
785 FOR_EACH_VEC_ELT (body, i, bb)
786 if (bb != entry_bb && bb != exit_bb)
787 for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi))
788 if (gimple_debug_bind_p (gsi_stmt (gsi)))
789 eliminate_local_variables_stmt (entry, &gsi, &decl_address);
792 /* Returns true if expression EXPR is not defined between ENTRY and
793 EXIT, i.e. if all its operands are defined outside of the region. */
795 static bool
796 expr_invariant_in_region_p (edge entry, edge exit, tree expr)
798 basic_block entry_bb = entry->src;
799 basic_block exit_bb = exit->dest;
800 basic_block def_bb;
802 if (is_gimple_min_invariant (expr))
803 return true;
805 if (TREE_CODE (expr) == SSA_NAME)
807 def_bb = gimple_bb (SSA_NAME_DEF_STMT (expr));
808 if (def_bb
809 && dominated_by_p (CDI_DOMINATORS, def_bb, entry_bb)
810 && !dominated_by_p (CDI_DOMINATORS, def_bb, exit_bb))
811 return false;
813 return true;
816 return false;
819 /* If COPY_NAME_P is true, creates and returns a duplicate of NAME.
820 The copies are stored to NAME_COPIES, if NAME was already duplicated,
821 its duplicate stored in NAME_COPIES is returned.
823 Regardless of COPY_NAME_P, the decl used as a base of the ssa name is also
824 duplicated, storing the copies in DECL_COPIES. */
826 static tree
827 separate_decls_in_region_name (tree name, name_to_copy_table_type *name_copies,
828 int_tree_htab_type *decl_copies,
829 bool copy_name_p)
831 tree copy, var, var_copy;
832 unsigned idx, uid, nuid;
833 struct int_tree_map ielt;
834 struct name_to_copy_elt elt, *nelt;
835 name_to_copy_elt **slot;
836 int_tree_map *dslot;
838 if (TREE_CODE (name) != SSA_NAME)
839 return name;
841 idx = SSA_NAME_VERSION (name);
842 elt.version = idx;
843 slot = name_copies->find_slot_with_hash (&elt, idx,
844 copy_name_p ? INSERT : NO_INSERT);
845 if (slot && *slot)
846 return (*slot)->new_name;
848 if (copy_name_p)
850 copy = duplicate_ssa_name (name, NULL);
851 nelt = XNEW (struct name_to_copy_elt);
852 nelt->version = idx;
853 nelt->new_name = copy;
854 nelt->field = NULL_TREE;
855 *slot = nelt;
857 else
859 gcc_assert (!slot);
860 copy = name;
863 var = SSA_NAME_VAR (name);
864 if (!var)
865 return copy;
867 uid = DECL_UID (var);
868 ielt.uid = uid;
869 dslot = decl_copies->find_slot_with_hash (ielt, uid, INSERT);
870 if (!dslot->to)
872 var_copy = create_tmp_var (TREE_TYPE (var), get_name (var));
873 DECL_GIMPLE_REG_P (var_copy) = DECL_GIMPLE_REG_P (var);
874 dslot->uid = uid;
875 dslot->to = var_copy;
877 /* Ensure that when we meet this decl next time, we won't duplicate
878 it again. */
879 nuid = DECL_UID (var_copy);
880 ielt.uid = nuid;
881 dslot = decl_copies->find_slot_with_hash (ielt, nuid, INSERT);
882 gcc_assert (!dslot->to);
883 dslot->uid = nuid;
884 dslot->to = var_copy;
886 else
887 var_copy = dslot->to;
889 replace_ssa_name_symbol (copy, var_copy);
890 return copy;
893 /* Finds the ssa names used in STMT that are defined outside the
894 region between ENTRY and EXIT and replaces such ssa names with
895 their duplicates. The duplicates are stored to NAME_COPIES. Base
896 decls of all ssa names used in STMT (including those defined in
897 LOOP) are replaced with the new temporary variables; the
898 replacement decls are stored in DECL_COPIES. */
900 static void
901 separate_decls_in_region_stmt (edge entry, edge exit, gimple *stmt,
902 name_to_copy_table_type *name_copies,
903 int_tree_htab_type *decl_copies)
905 use_operand_p use;
906 def_operand_p def;
907 ssa_op_iter oi;
908 tree name, copy;
909 bool copy_name_p;
911 FOR_EACH_PHI_OR_STMT_DEF (def, stmt, oi, SSA_OP_DEF)
913 name = DEF_FROM_PTR (def);
914 gcc_assert (TREE_CODE (name) == SSA_NAME);
915 copy = separate_decls_in_region_name (name, name_copies, decl_copies,
916 false);
917 gcc_assert (copy == name);
920 FOR_EACH_PHI_OR_STMT_USE (use, stmt, oi, SSA_OP_USE)
922 name = USE_FROM_PTR (use);
923 if (TREE_CODE (name) != SSA_NAME)
924 continue;
926 copy_name_p = expr_invariant_in_region_p (entry, exit, name);
927 copy = separate_decls_in_region_name (name, name_copies, decl_copies,
928 copy_name_p);
929 SET_USE (use, copy);
933 /* Finds the ssa names used in STMT that are defined outside the
934 region between ENTRY and EXIT and replaces such ssa names with
935 their duplicates. The duplicates are stored to NAME_COPIES. Base
936 decls of all ssa names used in STMT (including those defined in
937 LOOP) are replaced with the new temporary variables; the
938 replacement decls are stored in DECL_COPIES. */
940 static bool
941 separate_decls_in_region_debug (gimple *stmt,
942 name_to_copy_table_type *name_copies,
943 int_tree_htab_type *decl_copies)
945 use_operand_p use;
946 ssa_op_iter oi;
947 tree var, name;
948 struct int_tree_map ielt;
949 struct name_to_copy_elt elt;
950 name_to_copy_elt **slot;
951 int_tree_map *dslot;
953 if (gimple_debug_bind_p (stmt))
954 var = gimple_debug_bind_get_var (stmt);
955 else if (gimple_debug_source_bind_p (stmt))
956 var = gimple_debug_source_bind_get_var (stmt);
957 else
958 return true;
959 if (TREE_CODE (var) == DEBUG_EXPR_DECL || TREE_CODE (var) == LABEL_DECL)
960 return true;
961 gcc_assert (DECL_P (var) && SSA_VAR_P (var));
962 ielt.uid = DECL_UID (var);
963 dslot = decl_copies->find_slot_with_hash (ielt, ielt.uid, NO_INSERT);
964 if (!dslot)
965 return true;
966 if (gimple_debug_bind_p (stmt))
967 gimple_debug_bind_set_var (stmt, dslot->to);
968 else if (gimple_debug_source_bind_p (stmt))
969 gimple_debug_source_bind_set_var (stmt, dslot->to);
971 FOR_EACH_PHI_OR_STMT_USE (use, stmt, oi, SSA_OP_USE)
973 name = USE_FROM_PTR (use);
974 if (TREE_CODE (name) != SSA_NAME)
975 continue;
977 elt.version = SSA_NAME_VERSION (name);
978 slot = name_copies->find_slot_with_hash (&elt, elt.version, NO_INSERT);
979 if (!slot)
981 gimple_debug_bind_reset_value (stmt);
982 update_stmt (stmt);
983 break;
986 SET_USE (use, (*slot)->new_name);
989 return false;
992 /* Callback for htab_traverse. Adds a field corresponding to the reduction
993 specified in SLOT. The type is passed in DATA. */
996 add_field_for_reduction (reduction_info **slot, tree type)
999 struct reduction_info *const red = *slot;
1000 tree var = reduc_stmt_res (red->reduc_stmt);
1001 tree field = build_decl (gimple_location (red->reduc_stmt), FIELD_DECL,
1002 SSA_NAME_IDENTIFIER (var), TREE_TYPE (var));
1004 insert_field_into_struct (type, field);
1006 red->field = field;
1008 return 1;
1011 /* Callback for htab_traverse. Adds a field corresponding to a ssa name
1012 described in SLOT. The type is passed in DATA. */
1015 add_field_for_name (name_to_copy_elt **slot, tree type)
1017 struct name_to_copy_elt *const elt = *slot;
1018 tree name = ssa_name (elt->version);
1019 tree field = build_decl (UNKNOWN_LOCATION,
1020 FIELD_DECL, SSA_NAME_IDENTIFIER (name),
1021 TREE_TYPE (name));
1023 insert_field_into_struct (type, field);
1024 elt->field = field;
1026 return 1;
1029 /* Callback for htab_traverse. A local result is the intermediate result
1030 computed by a single
1031 thread, or the initial value in case no iteration was executed.
1032 This function creates a phi node reflecting these values.
1033 The phi's result will be stored in NEW_PHI field of the
1034 reduction's data structure. */
1037 create_phi_for_local_result (reduction_info **slot, struct loop *loop)
1039 struct reduction_info *const reduc = *slot;
1040 edge e;
1041 gphi *new_phi;
1042 basic_block store_bb, continue_bb;
1043 tree local_res;
1044 source_location locus;
1046 /* STORE_BB is the block where the phi
1047 should be stored. It is the destination of the loop exit.
1048 (Find the fallthru edge from GIMPLE_OMP_CONTINUE). */
1049 continue_bb = single_pred (loop->latch);
1050 store_bb = FALLTHRU_EDGE (continue_bb)->dest;
1052 /* STORE_BB has two predecessors. One coming from the loop
1053 (the reduction's result is computed at the loop),
1054 and another coming from a block preceding the loop,
1055 when no iterations
1056 are executed (the initial value should be taken). */
1057 if (EDGE_PRED (store_bb, 0) == FALLTHRU_EDGE (continue_bb))
1058 e = EDGE_PRED (store_bb, 1);
1059 else
1060 e = EDGE_PRED (store_bb, 0);
1061 tree lhs = reduc_stmt_res (reduc->reduc_stmt);
1062 local_res = copy_ssa_name (lhs);
1063 locus = gimple_location (reduc->reduc_stmt);
1064 new_phi = create_phi_node (local_res, store_bb);
1065 add_phi_arg (new_phi, reduc->init, e, locus);
1066 add_phi_arg (new_phi, lhs, FALLTHRU_EDGE (continue_bb), locus);
1067 reduc->new_phi = new_phi;
1069 return 1;
1072 struct clsn_data
1074 tree store;
1075 tree load;
1077 basic_block store_bb;
1078 basic_block load_bb;
1081 /* Callback for htab_traverse. Create an atomic instruction for the
1082 reduction described in SLOT.
1083 DATA annotates the place in memory the atomic operation relates to,
1084 and the basic block it needs to be generated in. */
1087 create_call_for_reduction_1 (reduction_info **slot, struct clsn_data *clsn_data)
1089 struct reduction_info *const reduc = *slot;
1090 gimple_stmt_iterator gsi;
1091 tree type = TREE_TYPE (PHI_RESULT (reduc->reduc_phi));
1092 tree load_struct;
1093 basic_block bb;
1094 basic_block new_bb;
1095 edge e;
1096 tree t, addr, ref, x;
1097 tree tmp_load, name;
1098 gimple *load;
1100 if (reduc->reduc_addr == NULL_TREE)
1102 load_struct = build_simple_mem_ref (clsn_data->load);
1103 t = build3 (COMPONENT_REF, type, load_struct, reduc->field, NULL_TREE);
1105 addr = build_addr (t);
1107 else
1109 /* Set the address for the atomic store. */
1110 addr = reduc->reduc_addr;
1112 /* Remove the non-atomic store '*addr = sum'. */
1113 tree res = PHI_RESULT (reduc->keep_res);
1114 use_operand_p use_p;
1115 gimple *stmt;
1116 bool single_use_p = single_imm_use (res, &use_p, &stmt);
1117 gcc_assert (single_use_p);
1118 replace_uses_by (gimple_vdef (stmt),
1119 gimple_vuse (stmt));
1120 gimple_stmt_iterator gsi = gsi_for_stmt (stmt);
1121 gsi_remove (&gsi, true);
1124 /* Create phi node. */
1125 bb = clsn_data->load_bb;
1127 gsi = gsi_last_bb (bb);
1128 e = split_block (bb, gsi_stmt (gsi));
1129 new_bb = e->dest;
1131 tmp_load = create_tmp_var (TREE_TYPE (TREE_TYPE (addr)));
1132 tmp_load = make_ssa_name (tmp_load);
1133 load = gimple_build_omp_atomic_load (tmp_load, addr);
1134 SSA_NAME_DEF_STMT (tmp_load) = load;
1135 gsi = gsi_start_bb (new_bb);
1136 gsi_insert_after (&gsi, load, GSI_NEW_STMT);
1138 e = split_block (new_bb, load);
1139 new_bb = e->dest;
1140 gsi = gsi_start_bb (new_bb);
1141 ref = tmp_load;
1142 x = fold_build2 (reduc->reduction_code,
1143 TREE_TYPE (PHI_RESULT (reduc->new_phi)), ref,
1144 PHI_RESULT (reduc->new_phi));
1146 name = force_gimple_operand_gsi (&gsi, x, true, NULL_TREE, true,
1147 GSI_CONTINUE_LINKING);
1149 gsi_insert_after (&gsi, gimple_build_omp_atomic_store (name), GSI_NEW_STMT);
1150 return 1;
1153 /* Create the atomic operation at the join point of the threads.
1154 REDUCTION_LIST describes the reductions in the LOOP.
1155 LD_ST_DATA describes the shared data structure where
1156 shared data is stored in and loaded from. */
1157 static void
1158 create_call_for_reduction (struct loop *loop,
1159 reduction_info_table_type *reduction_list,
1160 struct clsn_data *ld_st_data)
1162 reduction_list->traverse <struct loop *, create_phi_for_local_result> (loop);
1163 /* Find the fallthru edge from GIMPLE_OMP_CONTINUE. */
1164 basic_block continue_bb = single_pred (loop->latch);
1165 ld_st_data->load_bb = FALLTHRU_EDGE (continue_bb)->dest;
1166 reduction_list
1167 ->traverse <struct clsn_data *, create_call_for_reduction_1> (ld_st_data);
1170 /* Callback for htab_traverse. Loads the final reduction value at the
1171 join point of all threads, and inserts it in the right place. */
1174 create_loads_for_reductions (reduction_info **slot, struct clsn_data *clsn_data)
1176 struct reduction_info *const red = *slot;
1177 gimple *stmt;
1178 gimple_stmt_iterator gsi;
1179 tree type = TREE_TYPE (reduc_stmt_res (red->reduc_stmt));
1180 tree load_struct;
1181 tree name;
1182 tree x;
1184 /* If there's no exit phi, the result of the reduction is unused. */
1185 if (red->keep_res == NULL)
1186 return 1;
1188 gsi = gsi_after_labels (clsn_data->load_bb);
1189 load_struct = build_simple_mem_ref (clsn_data->load);
1190 load_struct = build3 (COMPONENT_REF, type, load_struct, red->field,
1191 NULL_TREE);
1193 x = load_struct;
1194 name = PHI_RESULT (red->keep_res);
1195 stmt = gimple_build_assign (name, x);
1197 gsi_insert_after (&gsi, stmt, GSI_NEW_STMT);
1199 for (gsi = gsi_start_phis (gimple_bb (red->keep_res));
1200 !gsi_end_p (gsi); gsi_next (&gsi))
1201 if (gsi_stmt (gsi) == red->keep_res)
1203 remove_phi_node (&gsi, false);
1204 return 1;
1206 gcc_unreachable ();
1209 /* Load the reduction result that was stored in LD_ST_DATA.
1210 REDUCTION_LIST describes the list of reductions that the
1211 loads should be generated for. */
1212 static void
1213 create_final_loads_for_reduction (reduction_info_table_type *reduction_list,
1214 struct clsn_data *ld_st_data)
1216 gimple_stmt_iterator gsi;
1217 tree t;
1218 gimple *stmt;
1220 gsi = gsi_after_labels (ld_st_data->load_bb);
1221 t = build_fold_addr_expr (ld_st_data->store);
1222 stmt = gimple_build_assign (ld_st_data->load, t);
1224 gsi_insert_before (&gsi, stmt, GSI_NEW_STMT);
1226 reduction_list
1227 ->traverse <struct clsn_data *, create_loads_for_reductions> (ld_st_data);
1231 /* Callback for htab_traverse. Store the neutral value for the
1232 particular reduction's operation, e.g. 0 for PLUS_EXPR,
1233 1 for MULT_EXPR, etc. into the reduction field.
1234 The reduction is specified in SLOT. The store information is
1235 passed in DATA. */
1238 create_stores_for_reduction (reduction_info **slot, struct clsn_data *clsn_data)
1240 struct reduction_info *const red = *slot;
1241 tree t;
1242 gimple *stmt;
1243 gimple_stmt_iterator gsi;
1244 tree type = TREE_TYPE (reduc_stmt_res (red->reduc_stmt));
1246 gsi = gsi_last_bb (clsn_data->store_bb);
1247 t = build3 (COMPONENT_REF, type, clsn_data->store, red->field, NULL_TREE);
1248 stmt = gimple_build_assign (t, red->initial_value);
1249 gsi_insert_after (&gsi, stmt, GSI_NEW_STMT);
1251 return 1;
1254 /* Callback for htab_traverse. Creates loads to a field of LOAD in LOAD_BB and
1255 store to a field of STORE in STORE_BB for the ssa name and its duplicate
1256 specified in SLOT. */
1259 create_loads_and_stores_for_name (name_to_copy_elt **slot,
1260 struct clsn_data *clsn_data)
1262 struct name_to_copy_elt *const elt = *slot;
1263 tree t;
1264 gimple *stmt;
1265 gimple_stmt_iterator gsi;
1266 tree type = TREE_TYPE (elt->new_name);
1267 tree load_struct;
1269 gsi = gsi_last_bb (clsn_data->store_bb);
1270 t = build3 (COMPONENT_REF, type, clsn_data->store, elt->field, NULL_TREE);
1271 stmt = gimple_build_assign (t, ssa_name (elt->version));
1272 gsi_insert_after (&gsi, stmt, GSI_NEW_STMT);
1274 gsi = gsi_last_bb (clsn_data->load_bb);
1275 load_struct = build_simple_mem_ref (clsn_data->load);
1276 t = build3 (COMPONENT_REF, type, load_struct, elt->field, NULL_TREE);
1277 stmt = gimple_build_assign (elt->new_name, t);
1278 gsi_insert_after (&gsi, stmt, GSI_NEW_STMT);
1280 return 1;
1283 /* Moves all the variables used in LOOP and defined outside of it (including
1284 the initial values of loop phi nodes, and *PER_THREAD if it is a ssa
1285 name) to a structure created for this purpose. The code
1287 while (1)
1289 use (a);
1290 use (b);
1293 is transformed this way:
1295 bb0:
1296 old.a = a;
1297 old.b = b;
1299 bb1:
1300 a' = new->a;
1301 b' = new->b;
1302 while (1)
1304 use (a');
1305 use (b');
1308 `old' is stored to *ARG_STRUCT and `new' is stored to NEW_ARG_STRUCT. The
1309 pointer `new' is intentionally not initialized (the loop will be split to a
1310 separate function later, and `new' will be initialized from its arguments).
1311 LD_ST_DATA holds information about the shared data structure used to pass
1312 information among the threads. It is initialized here, and
1313 gen_parallel_loop will pass it to create_call_for_reduction that
1314 needs this information. REDUCTION_LIST describes the reductions
1315 in LOOP. */
1317 static void
1318 separate_decls_in_region (edge entry, edge exit,
1319 reduction_info_table_type *reduction_list,
1320 tree *arg_struct, tree *new_arg_struct,
1321 struct clsn_data *ld_st_data)
1324 basic_block bb1 = split_edge (entry);
1325 basic_block bb0 = single_pred (bb1);
1326 name_to_copy_table_type name_copies (10);
1327 int_tree_htab_type decl_copies (10);
1328 unsigned i;
1329 tree type, type_name, nvar;
1330 gimple_stmt_iterator gsi;
1331 struct clsn_data clsn_data;
1332 auto_vec<basic_block, 3> body;
1333 basic_block bb;
1334 basic_block entry_bb = bb1;
1335 basic_block exit_bb = exit->dest;
1336 bool has_debug_stmt = false;
1338 entry = single_succ_edge (entry_bb);
1339 gather_blocks_in_sese_region (entry_bb, exit_bb, &body);
1341 FOR_EACH_VEC_ELT (body, i, bb)
1343 if (bb != entry_bb && bb != exit_bb)
1345 for (gsi = gsi_start_phis (bb); !gsi_end_p (gsi); gsi_next (&gsi))
1346 separate_decls_in_region_stmt (entry, exit, gsi_stmt (gsi),
1347 &name_copies, &decl_copies);
1349 for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi))
1351 gimple *stmt = gsi_stmt (gsi);
1353 if (is_gimple_debug (stmt))
1354 has_debug_stmt = true;
1355 else
1356 separate_decls_in_region_stmt (entry, exit, stmt,
1357 &name_copies, &decl_copies);
1362 /* Now process debug bind stmts. We must not create decls while
1363 processing debug stmts, so we defer their processing so as to
1364 make sure we will have debug info for as many variables as
1365 possible (all of those that were dealt with in the loop above),
1366 and discard those for which we know there's nothing we can
1367 do. */
1368 if (has_debug_stmt)
1369 FOR_EACH_VEC_ELT (body, i, bb)
1370 if (bb != entry_bb && bb != exit_bb)
1372 for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi);)
1374 gimple *stmt = gsi_stmt (gsi);
1376 if (is_gimple_debug (stmt))
1378 if (separate_decls_in_region_debug (stmt, &name_copies,
1379 &decl_copies))
1381 gsi_remove (&gsi, true);
1382 continue;
1386 gsi_next (&gsi);
1390 if (name_copies.elements () == 0 && reduction_list->elements () == 0)
1392 /* It may happen that there is nothing to copy (if there are only
1393 loop carried and external variables in the loop). */
1394 *arg_struct = NULL;
1395 *new_arg_struct = NULL;
1397 else
1399 /* Create the type for the structure to store the ssa names to. */
1400 type = lang_hooks.types.make_type (RECORD_TYPE);
1401 type_name = build_decl (UNKNOWN_LOCATION,
1402 TYPE_DECL, create_tmp_var_name (".paral_data"),
1403 type);
1404 TYPE_NAME (type) = type_name;
1406 name_copies.traverse <tree, add_field_for_name> (type);
1407 if (reduction_list && reduction_list->elements () > 0)
1409 /* Create the fields for reductions. */
1410 reduction_list->traverse <tree, add_field_for_reduction> (type);
1412 layout_type (type);
1414 /* Create the loads and stores. */
1415 *arg_struct = create_tmp_var (type, ".paral_data_store");
1416 nvar = create_tmp_var (build_pointer_type (type), ".paral_data_load");
1417 *new_arg_struct = make_ssa_name (nvar);
1419 ld_st_data->store = *arg_struct;
1420 ld_st_data->load = *new_arg_struct;
1421 ld_st_data->store_bb = bb0;
1422 ld_st_data->load_bb = bb1;
1424 name_copies
1425 .traverse <struct clsn_data *, create_loads_and_stores_for_name>
1426 (ld_st_data);
1428 /* Load the calculation from memory (after the join of the threads). */
1430 if (reduction_list && reduction_list->elements () > 0)
1432 reduction_list
1433 ->traverse <struct clsn_data *, create_stores_for_reduction>
1434 (ld_st_data);
1435 clsn_data.load = make_ssa_name (nvar);
1436 clsn_data.load_bb = exit->dest;
1437 clsn_data.store = ld_st_data->store;
1438 create_final_loads_for_reduction (reduction_list, &clsn_data);
1443 /* Returns true if FN was created to run in parallel. */
1445 bool
1446 parallelized_function_p (tree fndecl)
1448 cgraph_node *node = cgraph_node::get (fndecl);
1449 gcc_assert (node != NULL);
1450 return node->parallelized_function;
1453 /* Creates and returns an empty function that will receive the body of
1454 a parallelized loop. */
1456 static tree
1457 create_loop_fn (location_t loc)
1459 char buf[100];
1460 char *tname;
1461 tree decl, type, name, t;
1462 struct function *act_cfun = cfun;
1463 static unsigned loopfn_num;
1465 loc = LOCATION_LOCUS (loc);
1466 snprintf (buf, 100, "%s.$loopfn", current_function_name ());
1467 ASM_FORMAT_PRIVATE_NAME (tname, buf, loopfn_num++);
1468 clean_symbol_name (tname);
1469 name = get_identifier (tname);
1470 type = build_function_type_list (void_type_node, ptr_type_node, NULL_TREE);
1472 decl = build_decl (loc, FUNCTION_DECL, name, type);
1473 TREE_STATIC (decl) = 1;
1474 TREE_USED (decl) = 1;
1475 DECL_ARTIFICIAL (decl) = 1;
1476 DECL_IGNORED_P (decl) = 0;
1477 TREE_PUBLIC (decl) = 0;
1478 DECL_UNINLINABLE (decl) = 1;
1479 DECL_EXTERNAL (decl) = 0;
1480 DECL_CONTEXT (decl) = NULL_TREE;
1481 DECL_INITIAL (decl) = make_node (BLOCK);
1482 BLOCK_SUPERCONTEXT (DECL_INITIAL (decl)) = decl;
1484 t = build_decl (loc, RESULT_DECL, NULL_TREE, void_type_node);
1485 DECL_ARTIFICIAL (t) = 1;
1486 DECL_IGNORED_P (t) = 1;
1487 DECL_RESULT (decl) = t;
1489 t = build_decl (loc, PARM_DECL, get_identifier (".paral_data_param"),
1490 ptr_type_node);
1491 DECL_ARTIFICIAL (t) = 1;
1492 DECL_ARG_TYPE (t) = ptr_type_node;
1493 DECL_CONTEXT (t) = decl;
1494 TREE_USED (t) = 1;
1495 DECL_ARGUMENTS (decl) = t;
1497 allocate_struct_function (decl, false);
1499 /* The call to allocate_struct_function clobbers CFUN, so we need to restore
1500 it. */
1501 set_cfun (act_cfun);
1503 return decl;
1506 /* Replace uses of NAME by VAL in block BB. */
1508 static void
1509 replace_uses_in_bb_by (tree name, tree val, basic_block bb)
1511 gimple *use_stmt;
1512 imm_use_iterator imm_iter;
1514 FOR_EACH_IMM_USE_STMT (use_stmt, imm_iter, name)
1516 if (gimple_bb (use_stmt) != bb)
1517 continue;
1519 use_operand_p use_p;
1520 FOR_EACH_IMM_USE_ON_STMT (use_p, imm_iter)
1521 SET_USE (use_p, val);
1525 /* Do transformation from:
1527 <bb preheader>:
1529 goto <bb header>
1531 <bb header>:
1532 ivtmp_a = PHI <ivtmp_init (preheader), ivtmp_b (latch)>
1533 sum_a = PHI <sum_init (preheader), sum_b (latch)>
1535 use (ivtmp_a)
1537 sum_b = sum_a + sum_update
1539 if (ivtmp_a < n)
1540 goto <bb latch>;
1541 else
1542 goto <bb exit>;
1544 <bb latch>:
1545 ivtmp_b = ivtmp_a + 1;
1546 goto <bb header>
1548 <bb exit>:
1549 sum_z = PHI <sum_b (cond[1]), ...>
1551 [1] Where <bb cond> is single_pred (bb latch); In the simplest case,
1552 that's <bb header>.
1556 <bb preheader>:
1558 goto <bb newheader>
1560 <bb header>:
1561 ivtmp_a = PHI <ivtmp_c (latch)>
1562 sum_a = PHI <sum_c (latch)>
1564 use (ivtmp_a)
1566 sum_b = sum_a + sum_update
1568 goto <bb latch>;
1570 <bb newheader>:
1571 ivtmp_c = PHI <ivtmp_init (preheader), ivtmp_b (latch)>
1572 sum_c = PHI <sum_init (preheader), sum_b (latch)>
1573 if (ivtmp_c < n + 1)
1574 goto <bb header>;
1575 else
1576 goto <bb newexit>;
1578 <bb latch>:
1579 ivtmp_b = ivtmp_a + 1;
1580 goto <bb newheader>
1582 <bb newexit>:
1583 sum_y = PHI <sum_c (newheader)>
1585 <bb exit>:
1586 sum_z = PHI <sum_y (newexit), ...>
1589 In unified diff format:
1591 <bb preheader>:
1593 - goto <bb header>
1594 + goto <bb newheader>
1596 <bb header>:
1597 - ivtmp_a = PHI <ivtmp_init (preheader), ivtmp_b (latch)>
1598 - sum_a = PHI <sum_init (preheader), sum_b (latch)>
1599 + ivtmp_a = PHI <ivtmp_c (latch)>
1600 + sum_a = PHI <sum_c (latch)>
1602 use (ivtmp_a)
1604 sum_b = sum_a + sum_update
1606 - if (ivtmp_a < n)
1607 - goto <bb latch>;
1608 + goto <bb latch>;
1610 + <bb newheader>:
1611 + ivtmp_c = PHI <ivtmp_init (preheader), ivtmp_b (latch)>
1612 + sum_c = PHI <sum_init (preheader), sum_b (latch)>
1613 + if (ivtmp_c < n + 1)
1614 + goto <bb header>;
1615 else
1616 goto <bb exit>;
1618 <bb latch>:
1619 ivtmp_b = ivtmp_a + 1;
1620 - goto <bb header>
1621 + goto <bb newheader>
1623 + <bb newexit>:
1624 + sum_y = PHI <sum_c (newheader)>
1626 <bb exit>:
1627 - sum_z = PHI <sum_b (cond[1]), ...>
1628 + sum_z = PHI <sum_y (newexit), ...>
1630 Note: the example does not show any virtual phis, but these are handled more
1631 or less as reductions.
1634 Moves the exit condition of LOOP to the beginning of its header.
1635 REDUCTION_LIST describes the reductions in LOOP. BOUND is the new loop
1636 bound. */
1638 static void
1639 transform_to_exit_first_loop_alt (struct loop *loop,
1640 reduction_info_table_type *reduction_list,
1641 tree bound)
1643 basic_block header = loop->header;
1644 basic_block latch = loop->latch;
1645 edge exit = single_dom_exit (loop);
1646 basic_block exit_block = exit->dest;
1647 gcond *cond_stmt = as_a <gcond *> (last_stmt (exit->src));
1648 tree control = gimple_cond_lhs (cond_stmt);
1649 edge e;
1651 /* Rewriting virtuals into loop-closed ssa normal form makes this
1652 transformation simpler. It also ensures that the virtuals are in
1653 loop-closed ssa normal from after the transformation, which is required by
1654 create_parallel_loop. */
1655 rewrite_virtuals_into_loop_closed_ssa (loop);
1657 /* Create the new_header block. */
1658 basic_block new_header = split_block_before_cond_jump (exit->src);
1659 edge edge_at_split = single_pred_edge (new_header);
1661 /* Redirect entry edge to new_header. */
1662 edge entry = loop_preheader_edge (loop);
1663 e = redirect_edge_and_branch (entry, new_header);
1664 gcc_assert (e == entry);
1666 /* Redirect post_inc_edge to new_header. */
1667 edge post_inc_edge = single_succ_edge (latch);
1668 e = redirect_edge_and_branch (post_inc_edge, new_header);
1669 gcc_assert (e == post_inc_edge);
1671 /* Redirect post_cond_edge to header. */
1672 edge post_cond_edge = single_pred_edge (latch);
1673 e = redirect_edge_and_branch (post_cond_edge, header);
1674 gcc_assert (e == post_cond_edge);
1676 /* Redirect edge_at_split to latch. */
1677 e = redirect_edge_and_branch (edge_at_split, latch);
1678 gcc_assert (e == edge_at_split);
1680 /* Set the new loop bound. */
1681 gimple_cond_set_rhs (cond_stmt, bound);
1682 update_stmt (cond_stmt);
1684 /* Repair the ssa. */
1685 vec<edge_var_map> *v = redirect_edge_var_map_vector (post_inc_edge);
1686 edge_var_map *vm;
1687 gphi_iterator gsi;
1688 int i;
1689 for (gsi = gsi_start_phis (header), i = 0;
1690 !gsi_end_p (gsi) && v->iterate (i, &vm);
1691 gsi_next (&gsi), i++)
1693 gphi *phi = gsi.phi ();
1694 tree res_a = PHI_RESULT (phi);
1696 /* Create new phi. */
1697 tree res_c = copy_ssa_name (res_a, phi);
1698 gphi *nphi = create_phi_node (res_c, new_header);
1700 /* Replace ivtmp_a with ivtmp_c in condition 'if (ivtmp_a < n)'. */
1701 replace_uses_in_bb_by (res_a, res_c, new_header);
1703 /* Replace ivtmp/sum_b with ivtmp/sum_c in header phi. */
1704 add_phi_arg (phi, res_c, post_cond_edge, UNKNOWN_LOCATION);
1706 /* Replace sum_b with sum_c in exit phi. */
1707 tree res_b = redirect_edge_var_map_def (vm);
1708 replace_uses_in_bb_by (res_b, res_c, exit_block);
1710 struct reduction_info *red = reduction_phi (reduction_list, phi);
1711 gcc_assert (virtual_operand_p (res_a)
1712 || res_a == control
1713 || red != NULL);
1715 if (red)
1717 /* Register the new reduction phi. */
1718 red->reduc_phi = nphi;
1719 gimple_set_uid (red->reduc_phi, red->reduc_version);
1722 gcc_assert (gsi_end_p (gsi) && !v->iterate (i, &vm));
1724 /* Set the preheader argument of the new phis to ivtmp/sum_init. */
1725 flush_pending_stmts (entry);
1727 /* Set the latch arguments of the new phis to ivtmp/sum_b. */
1728 flush_pending_stmts (post_inc_edge);
1731 basic_block new_exit_block = NULL;
1732 if (!single_pred_p (exit->dest))
1734 /* Create a new empty exit block, inbetween the new loop header and the
1735 old exit block. The function separate_decls_in_region needs this block
1736 to insert code that is active on loop exit, but not any other path. */
1737 new_exit_block = split_edge (exit);
1740 /* Insert and register the reduction exit phis. */
1741 for (gphi_iterator gsi = gsi_start_phis (exit_block);
1742 !gsi_end_p (gsi);
1743 gsi_next (&gsi))
1745 gphi *phi = gsi.phi ();
1746 gphi *nphi = NULL;
1747 tree res_z = PHI_RESULT (phi);
1748 tree res_c;
1750 if (new_exit_block != NULL)
1752 /* Now that we have a new exit block, duplicate the phi of the old
1753 exit block in the new exit block to preserve loop-closed ssa. */
1754 edge succ_new_exit_block = single_succ_edge (new_exit_block);
1755 edge pred_new_exit_block = single_pred_edge (new_exit_block);
1756 tree res_y = copy_ssa_name (res_z, phi);
1757 nphi = create_phi_node (res_y, new_exit_block);
1758 res_c = PHI_ARG_DEF_FROM_EDGE (phi, succ_new_exit_block);
1759 add_phi_arg (nphi, res_c, pred_new_exit_block, UNKNOWN_LOCATION);
1760 add_phi_arg (phi, res_y, succ_new_exit_block, UNKNOWN_LOCATION);
1762 else
1763 res_c = PHI_ARG_DEF_FROM_EDGE (phi, exit);
1765 if (virtual_operand_p (res_z))
1766 continue;
1768 gimple *reduc_phi = SSA_NAME_DEF_STMT (res_c);
1769 struct reduction_info *red = reduction_phi (reduction_list, reduc_phi);
1770 if (red != NULL)
1771 red->keep_res = (nphi != NULL
1772 ? nphi
1773 : phi);
1776 /* We're going to cancel the loop at the end of gen_parallel_loop, but until
1777 then we're still using some fields, so only bother about fields that are
1778 still used: header and latch.
1779 The loop has a new header bb, so we update it. The latch bb stays the
1780 same. */
1781 loop->header = new_header;
1783 /* Recalculate dominance info. */
1784 free_dominance_info (CDI_DOMINATORS);
1785 calculate_dominance_info (CDI_DOMINATORS);
1787 checking_verify_ssa (true, true);
1790 /* Tries to moves the exit condition of LOOP to the beginning of its header
1791 without duplication of the loop body. NIT is the number of iterations of the
1792 loop. REDUCTION_LIST describes the reductions in LOOP. Return true if
1793 transformation is successful. */
1795 static bool
1796 try_transform_to_exit_first_loop_alt (struct loop *loop,
1797 reduction_info_table_type *reduction_list,
1798 tree nit)
1800 /* Check whether the latch contains a single statement. */
1801 if (!gimple_seq_nondebug_singleton_p (bb_seq (loop->latch)))
1802 return false;
1804 /* Check whether the latch contains no phis. */
1805 if (phi_nodes (loop->latch) != NULL)
1806 return false;
1808 /* Check whether the latch contains the loop iv increment. */
1809 edge back = single_succ_edge (loop->latch);
1810 edge exit = single_dom_exit (loop);
1811 gcond *cond_stmt = as_a <gcond *> (last_stmt (exit->src));
1812 tree control = gimple_cond_lhs (cond_stmt);
1813 gphi *phi = as_a <gphi *> (SSA_NAME_DEF_STMT (control));
1814 tree inc_res = gimple_phi_arg_def (phi, back->dest_idx);
1815 if (gimple_bb (SSA_NAME_DEF_STMT (inc_res)) != loop->latch)
1816 return false;
1818 /* Check whether there's no code between the loop condition and the latch. */
1819 if (!single_pred_p (loop->latch)
1820 || single_pred (loop->latch) != exit->src)
1821 return false;
1823 tree alt_bound = NULL_TREE;
1824 tree nit_type = TREE_TYPE (nit);
1826 /* Figure out whether nit + 1 overflows. */
1827 if (TREE_CODE (nit) == INTEGER_CST)
1829 if (!tree_int_cst_equal (nit, TYPE_MAX_VALUE (nit_type)))
1831 alt_bound = fold_build2_loc (UNKNOWN_LOCATION, PLUS_EXPR, nit_type,
1832 nit, build_one_cst (nit_type));
1834 gcc_assert (TREE_CODE (alt_bound) == INTEGER_CST);
1835 transform_to_exit_first_loop_alt (loop, reduction_list, alt_bound);
1836 return true;
1838 else
1840 /* Todo: Figure out if we can trigger this, if it's worth to handle
1841 optimally, and if we can handle it optimally. */
1842 return false;
1846 gcc_assert (TREE_CODE (nit) == SSA_NAME);
1848 /* Variable nit is the loop bound as returned by canonicalize_loop_ivs, for an
1849 iv with base 0 and step 1 that is incremented in the latch, like this:
1851 <bb header>:
1852 # iv_1 = PHI <0 (preheader), iv_2 (latch)>
1854 if (iv_1 < nit)
1855 goto <bb latch>;
1856 else
1857 goto <bb exit>;
1859 <bb latch>:
1860 iv_2 = iv_1 + 1;
1861 goto <bb header>;
1863 The range of iv_1 is [0, nit]. The latch edge is taken for
1864 iv_1 == [0, nit - 1] and the exit edge is taken for iv_1 == nit. So the
1865 number of latch executions is equal to nit.
1867 The function max_loop_iterations gives us the maximum number of latch
1868 executions, so it gives us the maximum value of nit. */
1869 widest_int nit_max;
1870 if (!max_loop_iterations (loop, &nit_max))
1871 return false;
1873 /* Check if nit + 1 overflows. */
1874 widest_int type_max = wi::to_widest (TYPE_MAX_VALUE (nit_type));
1875 if (nit_max >= type_max)
1876 return false;
1878 gimple *def = SSA_NAME_DEF_STMT (nit);
1880 /* Try to find nit + 1, in the form of n in an assignment nit = n - 1. */
1881 if (def
1882 && is_gimple_assign (def)
1883 && gimple_assign_rhs_code (def) == PLUS_EXPR)
1885 tree op1 = gimple_assign_rhs1 (def);
1886 tree op2 = gimple_assign_rhs2 (def);
1887 if (integer_minus_onep (op1))
1888 alt_bound = op2;
1889 else if (integer_minus_onep (op2))
1890 alt_bound = op1;
1893 /* If not found, insert nit + 1. */
1894 if (alt_bound == NULL_TREE)
1896 alt_bound = fold_build2 (PLUS_EXPR, nit_type, nit,
1897 build_int_cst_type (nit_type, 1));
1899 gimple_stmt_iterator gsi = gsi_last_bb (loop_preheader_edge (loop)->src);
1901 alt_bound
1902 = force_gimple_operand_gsi (&gsi, alt_bound, true, NULL_TREE, false,
1903 GSI_CONTINUE_LINKING);
1906 transform_to_exit_first_loop_alt (loop, reduction_list, alt_bound);
1907 return true;
1910 /* Moves the exit condition of LOOP to the beginning of its header. NIT is the
1911 number of iterations of the loop. REDUCTION_LIST describes the reductions in
1912 LOOP. */
1914 static void
1915 transform_to_exit_first_loop (struct loop *loop,
1916 reduction_info_table_type *reduction_list,
1917 tree nit)
1919 basic_block *bbs, *nbbs, ex_bb, orig_header;
1920 unsigned n;
1921 bool ok;
1922 edge exit = single_dom_exit (loop), hpred;
1923 tree control, control_name, res, t;
1924 gphi *phi, *nphi;
1925 gassign *stmt;
1926 gcond *cond_stmt, *cond_nit;
1927 tree nit_1;
1929 split_block_after_labels (loop->header);
1930 orig_header = single_succ (loop->header);
1931 hpred = single_succ_edge (loop->header);
1933 cond_stmt = as_a <gcond *> (last_stmt (exit->src));
1934 control = gimple_cond_lhs (cond_stmt);
1935 gcc_assert (gimple_cond_rhs (cond_stmt) == nit);
1937 /* Make sure that we have phi nodes on exit for all loop header phis
1938 (create_parallel_loop requires that). */
1939 for (gphi_iterator gsi = gsi_start_phis (loop->header);
1940 !gsi_end_p (gsi);
1941 gsi_next (&gsi))
1943 phi = gsi.phi ();
1944 res = PHI_RESULT (phi);
1945 t = copy_ssa_name (res, phi);
1946 SET_PHI_RESULT (phi, t);
1947 nphi = create_phi_node (res, orig_header);
1948 add_phi_arg (nphi, t, hpred, UNKNOWN_LOCATION);
1950 if (res == control)
1952 gimple_cond_set_lhs (cond_stmt, t);
1953 update_stmt (cond_stmt);
1954 control = t;
1958 bbs = get_loop_body_in_dom_order (loop);
1960 for (n = 0; bbs[n] != exit->src; n++)
1961 continue;
1962 nbbs = XNEWVEC (basic_block, n);
1963 ok = gimple_duplicate_sese_tail (single_succ_edge (loop->header), exit,
1964 bbs + 1, n, nbbs);
1965 gcc_assert (ok);
1966 free (bbs);
1967 ex_bb = nbbs[0];
1968 free (nbbs);
1970 /* Other than reductions, the only gimple reg that should be copied
1971 out of the loop is the control variable. */
1972 exit = single_dom_exit (loop);
1973 control_name = NULL_TREE;
1974 for (gphi_iterator gsi = gsi_start_phis (ex_bb);
1975 !gsi_end_p (gsi); )
1977 phi = gsi.phi ();
1978 res = PHI_RESULT (phi);
1979 if (virtual_operand_p (res))
1981 gsi_next (&gsi);
1982 continue;
1985 /* Check if it is a part of reduction. If it is,
1986 keep the phi at the reduction's keep_res field. The
1987 PHI_RESULT of this phi is the resulting value of the reduction
1988 variable when exiting the loop. */
1990 if (reduction_list->elements () > 0)
1992 struct reduction_info *red;
1994 tree val = PHI_ARG_DEF_FROM_EDGE (phi, exit);
1995 red = reduction_phi (reduction_list, SSA_NAME_DEF_STMT (val));
1996 if (red)
1998 red->keep_res = phi;
1999 gsi_next (&gsi);
2000 continue;
2003 gcc_assert (control_name == NULL_TREE
2004 && SSA_NAME_VAR (res) == SSA_NAME_VAR (control));
2005 control_name = res;
2006 remove_phi_node (&gsi, false);
2008 gcc_assert (control_name != NULL_TREE);
2010 /* Initialize the control variable to number of iterations
2011 according to the rhs of the exit condition. */
2012 gimple_stmt_iterator gsi = gsi_after_labels (ex_bb);
2013 cond_nit = as_a <gcond *> (last_stmt (exit->src));
2014 nit_1 = gimple_cond_rhs (cond_nit);
2015 nit_1 = force_gimple_operand_gsi (&gsi,
2016 fold_convert (TREE_TYPE (control_name), nit_1),
2017 false, NULL_TREE, false, GSI_SAME_STMT);
2018 stmt = gimple_build_assign (control_name, nit_1);
2019 gsi_insert_before (&gsi, stmt, GSI_NEW_STMT);
2022 /* Create the parallel constructs for LOOP as described in gen_parallel_loop.
2023 LOOP_FN and DATA are the arguments of GIMPLE_OMP_PARALLEL.
2024 NEW_DATA is the variable that should be initialized from the argument
2025 of LOOP_FN. N_THREADS is the requested number of threads, which can be 0 if
2026 that number is to be determined later. */
2028 static void
2029 create_parallel_loop (struct loop *loop, tree loop_fn, tree data,
2030 tree new_data, unsigned n_threads, location_t loc,
2031 bool oacc_kernels_p)
2033 gimple_stmt_iterator gsi;
2034 basic_block for_bb, ex_bb, continue_bb;
2035 tree t, param;
2036 gomp_parallel *omp_par_stmt;
2037 gimple *omp_return_stmt1, *omp_return_stmt2;
2038 gimple *phi;
2039 gcond *cond_stmt;
2040 gomp_for *for_stmt;
2041 gomp_continue *omp_cont_stmt;
2042 tree cvar, cvar_init, initvar, cvar_next, cvar_base, type;
2043 edge exit, nexit, guard, end, e;
2045 if (oacc_kernels_p)
2047 gcc_checking_assert (lookup_attribute ("oacc kernels",
2048 DECL_ATTRIBUTES (cfun->decl)));
2049 /* Indicate to later processing that this is a parallelized OpenACC
2050 kernels construct. */
2051 DECL_ATTRIBUTES (cfun->decl)
2052 = tree_cons (get_identifier ("oacc kernels parallelized"),
2053 NULL_TREE, DECL_ATTRIBUTES (cfun->decl));
2055 else
2057 /* Prepare the GIMPLE_OMP_PARALLEL statement. */
2059 basic_block bb = loop_preheader_edge (loop)->src;
2060 basic_block paral_bb = single_pred (bb);
2061 gsi = gsi_last_bb (paral_bb);
2063 gcc_checking_assert (n_threads != 0);
2064 t = build_omp_clause (loc, OMP_CLAUSE_NUM_THREADS);
2065 OMP_CLAUSE_NUM_THREADS_EXPR (t)
2066 = build_int_cst (integer_type_node, n_threads);
2067 omp_par_stmt = gimple_build_omp_parallel (NULL, t, loop_fn, data);
2068 gimple_set_location (omp_par_stmt, loc);
2070 gsi_insert_after (&gsi, omp_par_stmt, GSI_NEW_STMT);
2072 /* Initialize NEW_DATA. */
2073 if (data)
2075 gassign *assign_stmt;
2077 gsi = gsi_after_labels (bb);
2079 param = make_ssa_name (DECL_ARGUMENTS (loop_fn));
2080 assign_stmt = gimple_build_assign (param, build_fold_addr_expr (data));
2081 gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT);
2083 assign_stmt = gimple_build_assign (new_data,
2084 fold_convert (TREE_TYPE (new_data), param));
2085 gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT);
2088 /* Emit GIMPLE_OMP_RETURN for GIMPLE_OMP_PARALLEL. */
2089 bb = split_loop_exit_edge (single_dom_exit (loop));
2090 gsi = gsi_last_bb (bb);
2091 omp_return_stmt1 = gimple_build_omp_return (false);
2092 gimple_set_location (omp_return_stmt1, loc);
2093 gsi_insert_after (&gsi, omp_return_stmt1, GSI_NEW_STMT);
2096 /* Extract data for GIMPLE_OMP_FOR. */
2097 gcc_assert (loop->header == single_dom_exit (loop)->src);
2098 cond_stmt = as_a <gcond *> (last_stmt (loop->header));
2100 cvar = gimple_cond_lhs (cond_stmt);
2101 cvar_base = SSA_NAME_VAR (cvar);
2102 phi = SSA_NAME_DEF_STMT (cvar);
2103 cvar_init = PHI_ARG_DEF_FROM_EDGE (phi, loop_preheader_edge (loop));
2104 initvar = copy_ssa_name (cvar);
2105 SET_USE (PHI_ARG_DEF_PTR_FROM_EDGE (phi, loop_preheader_edge (loop)),
2106 initvar);
2107 cvar_next = PHI_ARG_DEF_FROM_EDGE (phi, loop_latch_edge (loop));
2109 gsi = gsi_last_nondebug_bb (loop->latch);
2110 gcc_assert (gsi_stmt (gsi) == SSA_NAME_DEF_STMT (cvar_next));
2111 gsi_remove (&gsi, true);
2113 /* Prepare cfg. */
2114 for_bb = split_edge (loop_preheader_edge (loop));
2115 ex_bb = split_loop_exit_edge (single_dom_exit (loop));
2116 extract_true_false_edges_from_block (loop->header, &nexit, &exit);
2117 gcc_assert (exit == single_dom_exit (loop));
2119 guard = make_edge (for_bb, ex_bb, 0);
2120 /* FIXME: What is the probability? */
2121 guard->probability = profile_probability::guessed_never ();
2122 /* Split the latch edge, so LOOPS_HAVE_SIMPLE_LATCHES is still valid. */
2123 loop->latch = split_edge (single_succ_edge (loop->latch));
2124 single_pred_edge (loop->latch)->flags = 0;
2125 end = make_single_succ_edge (single_pred (loop->latch), ex_bb, EDGE_FALLTHRU);
2126 rescan_loop_exit (end, true, false);
2128 for (gphi_iterator gpi = gsi_start_phis (ex_bb);
2129 !gsi_end_p (gpi); gsi_next (&gpi))
2131 source_location locus;
2132 gphi *phi = gpi.phi ();
2133 tree def = PHI_ARG_DEF_FROM_EDGE (phi, exit);
2134 gimple *def_stmt = SSA_NAME_DEF_STMT (def);
2136 /* If the exit phi is not connected to a header phi in the same loop, this
2137 value is not modified in the loop, and we're done with this phi. */
2138 if (!(gimple_code (def_stmt) == GIMPLE_PHI
2139 && gimple_bb (def_stmt) == loop->header))
2141 locus = gimple_phi_arg_location_from_edge (phi, exit);
2142 add_phi_arg (phi, def, guard, locus);
2143 add_phi_arg (phi, def, end, locus);
2144 continue;
2147 gphi *stmt = as_a <gphi *> (def_stmt);
2148 def = PHI_ARG_DEF_FROM_EDGE (stmt, loop_preheader_edge (loop));
2149 locus = gimple_phi_arg_location_from_edge (stmt,
2150 loop_preheader_edge (loop));
2151 add_phi_arg (phi, def, guard, locus);
2153 def = PHI_ARG_DEF_FROM_EDGE (stmt, loop_latch_edge (loop));
2154 locus = gimple_phi_arg_location_from_edge (stmt, loop_latch_edge (loop));
2155 add_phi_arg (phi, def, end, locus);
2157 e = redirect_edge_and_branch (exit, nexit->dest);
2158 PENDING_STMT (e) = NULL;
2160 /* Emit GIMPLE_OMP_FOR. */
2161 if (oacc_kernels_p)
2162 /* Parallelized OpenACC kernels constructs use gang parallelism. See also
2163 omp-offload.c:execute_oacc_device_lower. */
2164 t = build_omp_clause (loc, OMP_CLAUSE_GANG);
2165 else
2167 t = build_omp_clause (loc, OMP_CLAUSE_SCHEDULE);
2168 int chunk_size = PARAM_VALUE (PARAM_PARLOOPS_CHUNK_SIZE);
2169 enum PARAM_PARLOOPS_SCHEDULE_KIND schedule_type \
2170 = (enum PARAM_PARLOOPS_SCHEDULE_KIND) PARAM_VALUE (PARAM_PARLOOPS_SCHEDULE);
2171 switch (schedule_type)
2173 case PARAM_PARLOOPS_SCHEDULE_KIND_static:
2174 OMP_CLAUSE_SCHEDULE_KIND (t) = OMP_CLAUSE_SCHEDULE_STATIC;
2175 break;
2176 case PARAM_PARLOOPS_SCHEDULE_KIND_dynamic:
2177 OMP_CLAUSE_SCHEDULE_KIND (t) = OMP_CLAUSE_SCHEDULE_DYNAMIC;
2178 break;
2179 case PARAM_PARLOOPS_SCHEDULE_KIND_guided:
2180 OMP_CLAUSE_SCHEDULE_KIND (t) = OMP_CLAUSE_SCHEDULE_GUIDED;
2181 break;
2182 case PARAM_PARLOOPS_SCHEDULE_KIND_auto:
2183 OMP_CLAUSE_SCHEDULE_KIND (t) = OMP_CLAUSE_SCHEDULE_AUTO;
2184 chunk_size = 0;
2185 break;
2186 case PARAM_PARLOOPS_SCHEDULE_KIND_runtime:
2187 OMP_CLAUSE_SCHEDULE_KIND (t) = OMP_CLAUSE_SCHEDULE_RUNTIME;
2188 chunk_size = 0;
2189 break;
2190 default:
2191 gcc_unreachable ();
2193 if (chunk_size != 0)
2194 OMP_CLAUSE_SCHEDULE_CHUNK_EXPR (t)
2195 = build_int_cst (integer_type_node, chunk_size);
2198 for_stmt = gimple_build_omp_for (NULL,
2199 (oacc_kernels_p
2200 ? GF_OMP_FOR_KIND_OACC_LOOP
2201 : GF_OMP_FOR_KIND_FOR),
2202 t, 1, NULL);
2204 gimple_cond_set_lhs (cond_stmt, cvar_base);
2205 type = TREE_TYPE (cvar);
2206 gimple_set_location (for_stmt, loc);
2207 gimple_omp_for_set_index (for_stmt, 0, initvar);
2208 gimple_omp_for_set_initial (for_stmt, 0, cvar_init);
2209 gimple_omp_for_set_final (for_stmt, 0, gimple_cond_rhs (cond_stmt));
2210 gimple_omp_for_set_cond (for_stmt, 0, gimple_cond_code (cond_stmt));
2211 gimple_omp_for_set_incr (for_stmt, 0, build2 (PLUS_EXPR, type,
2212 cvar_base,
2213 build_int_cst (type, 1)));
2215 gsi = gsi_last_bb (for_bb);
2216 gsi_insert_after (&gsi, for_stmt, GSI_NEW_STMT);
2217 SSA_NAME_DEF_STMT (initvar) = for_stmt;
2219 /* Emit GIMPLE_OMP_CONTINUE. */
2220 continue_bb = single_pred (loop->latch);
2221 gsi = gsi_last_bb (continue_bb);
2222 omp_cont_stmt = gimple_build_omp_continue (cvar_next, cvar);
2223 gimple_set_location (omp_cont_stmt, loc);
2224 gsi_insert_after (&gsi, omp_cont_stmt, GSI_NEW_STMT);
2225 SSA_NAME_DEF_STMT (cvar_next) = omp_cont_stmt;
2227 /* Emit GIMPLE_OMP_RETURN for GIMPLE_OMP_FOR. */
2228 gsi = gsi_last_bb (ex_bb);
2229 omp_return_stmt2 = gimple_build_omp_return (true);
2230 gimple_set_location (omp_return_stmt2, loc);
2231 gsi_insert_after (&gsi, omp_return_stmt2, GSI_NEW_STMT);
2233 /* After the above dom info is hosed. Re-compute it. */
2234 free_dominance_info (CDI_DOMINATORS);
2235 calculate_dominance_info (CDI_DOMINATORS);
2238 /* Return number of phis in bb. If COUNT_VIRTUAL_P is false, don't count the
2239 virtual phi. */
2241 static unsigned int
2242 num_phis (basic_block bb, bool count_virtual_p)
2244 unsigned int nr_phis = 0;
2245 gphi_iterator gsi;
2246 for (gsi = gsi_start_phis (bb); !gsi_end_p (gsi); gsi_next (&gsi))
2248 if (!count_virtual_p && virtual_operand_p (PHI_RESULT (gsi.phi ())))
2249 continue;
2251 nr_phis++;
2254 return nr_phis;
2257 /* Generates code to execute the iterations of LOOP in N_THREADS
2258 threads in parallel, which can be 0 if that number is to be determined
2259 later.
2261 NITER describes number of iterations of LOOP.
2262 REDUCTION_LIST describes the reductions existent in the LOOP. */
2264 static void
2265 gen_parallel_loop (struct loop *loop,
2266 reduction_info_table_type *reduction_list,
2267 unsigned n_threads, struct tree_niter_desc *niter,
2268 bool oacc_kernels_p)
2270 tree many_iterations_cond, type, nit;
2271 tree arg_struct, new_arg_struct;
2272 gimple_seq stmts;
2273 edge entry, exit;
2274 struct clsn_data clsn_data;
2275 location_t loc;
2276 gimple *cond_stmt;
2277 unsigned int m_p_thread=2;
2279 /* From
2281 ---------------------------------------------------------------------
2282 loop
2284 IV = phi (INIT, IV + STEP)
2285 BODY1;
2286 if (COND)
2287 break;
2288 BODY2;
2290 ---------------------------------------------------------------------
2292 with # of iterations NITER (possibly with MAY_BE_ZERO assumption),
2293 we generate the following code:
2295 ---------------------------------------------------------------------
2297 if (MAY_BE_ZERO
2298 || NITER < MIN_PER_THREAD * N_THREADS)
2299 goto original;
2301 BODY1;
2302 store all local loop-invariant variables used in body of the loop to DATA.
2303 GIMPLE_OMP_PARALLEL (OMP_CLAUSE_NUM_THREADS (N_THREADS), LOOPFN, DATA);
2304 load the variables from DATA.
2305 GIMPLE_OMP_FOR (IV = INIT; COND; IV += STEP) (OMP_CLAUSE_SCHEDULE (static))
2306 BODY2;
2307 BODY1;
2308 GIMPLE_OMP_CONTINUE;
2309 GIMPLE_OMP_RETURN -- GIMPLE_OMP_FOR
2310 GIMPLE_OMP_RETURN -- GIMPLE_OMP_PARALLEL
2311 goto end;
2313 original:
2314 loop
2316 IV = phi (INIT, IV + STEP)
2317 BODY1;
2318 if (COND)
2319 break;
2320 BODY2;
2323 end:
2327 /* Create two versions of the loop -- in the old one, we know that the
2328 number of iterations is large enough, and we will transform it into the
2329 loop that will be split to loop_fn, the new one will be used for the
2330 remaining iterations. */
2332 /* We should compute a better number-of-iterations value for outer loops.
2333 That is, if we have
2335 for (i = 0; i < n; ++i)
2336 for (j = 0; j < m; ++j)
2339 we should compute nit = n * m, not nit = n.
2340 Also may_be_zero handling would need to be adjusted. */
2342 type = TREE_TYPE (niter->niter);
2343 nit = force_gimple_operand (unshare_expr (niter->niter), &stmts, true,
2344 NULL_TREE);
2345 if (stmts)
2346 gsi_insert_seq_on_edge_immediate (loop_preheader_edge (loop), stmts);
2348 if (!oacc_kernels_p)
2350 if (loop->inner)
2351 m_p_thread=2;
2352 else
2353 m_p_thread=MIN_PER_THREAD;
2355 gcc_checking_assert (n_threads != 0);
2356 many_iterations_cond =
2357 fold_build2 (GE_EXPR, boolean_type_node,
2358 nit, build_int_cst (type, m_p_thread * n_threads - 1));
2360 many_iterations_cond
2361 = fold_build2 (TRUTH_AND_EXPR, boolean_type_node,
2362 invert_truthvalue (unshare_expr (niter->may_be_zero)),
2363 many_iterations_cond);
2364 many_iterations_cond
2365 = force_gimple_operand (many_iterations_cond, &stmts, false, NULL_TREE);
2366 if (stmts)
2367 gsi_insert_seq_on_edge_immediate (loop_preheader_edge (loop), stmts);
2368 if (!is_gimple_condexpr (many_iterations_cond))
2370 many_iterations_cond
2371 = force_gimple_operand (many_iterations_cond, &stmts,
2372 true, NULL_TREE);
2373 if (stmts)
2374 gsi_insert_seq_on_edge_immediate (loop_preheader_edge (loop),
2375 stmts);
2378 initialize_original_copy_tables ();
2380 /* We assume that the loop usually iterates a lot. */
2381 loop_version (loop, many_iterations_cond, NULL,
2382 profile_probability::likely (),
2383 profile_probability::unlikely (),
2384 profile_probability::likely (),
2385 profile_probability::unlikely (), true);
2386 update_ssa (TODO_update_ssa);
2387 free_original_copy_tables ();
2390 /* Base all the induction variables in LOOP on a single control one. */
2391 canonicalize_loop_ivs (loop, &nit, true);
2392 if (num_phis (loop->header, false) != reduction_list->elements () + 1)
2394 /* The call to canonicalize_loop_ivs above failed to "base all the
2395 induction variables in LOOP on a single control one". Do damage
2396 control. */
2397 basic_block preheader = loop_preheader_edge (loop)->src;
2398 basic_block cond_bb = single_pred (preheader);
2399 gcond *cond = as_a <gcond *> (gsi_stmt (gsi_last_bb (cond_bb)));
2400 gimple_cond_make_true (cond);
2401 update_stmt (cond);
2402 /* We've gotten rid of the duplicate loop created by loop_version, but
2403 we can't undo whatever canonicalize_loop_ivs has done.
2404 TODO: Fix this properly by ensuring that the call to
2405 canonicalize_loop_ivs succeeds. */
2406 if (dump_file
2407 && (dump_flags & TDF_DETAILS))
2408 fprintf (dump_file, "canonicalize_loop_ivs failed for loop %d,"
2409 " aborting transformation\n", loop->num);
2410 return;
2413 /* Ensure that the exit condition is the first statement in the loop.
2414 The common case is that latch of the loop is empty (apart from the
2415 increment) and immediately follows the loop exit test. Attempt to move the
2416 entry of the loop directly before the exit check and increase the number of
2417 iterations of the loop by one. */
2418 if (try_transform_to_exit_first_loop_alt (loop, reduction_list, nit))
2420 if (dump_file
2421 && (dump_flags & TDF_DETAILS))
2422 fprintf (dump_file,
2423 "alternative exit-first loop transform succeeded"
2424 " for loop %d\n", loop->num);
2426 else
2428 if (oacc_kernels_p)
2429 n_threads = 1;
2431 /* Fall back on the method that handles more cases, but duplicates the
2432 loop body: move the exit condition of LOOP to the beginning of its
2433 header, and duplicate the part of the last iteration that gets disabled
2434 to the exit of the loop. */
2435 transform_to_exit_first_loop (loop, reduction_list, nit);
2438 /* Generate initializations for reductions. */
2439 if (reduction_list->elements () > 0)
2440 reduction_list->traverse <struct loop *, initialize_reductions> (loop);
2442 /* Eliminate the references to local variables from the loop. */
2443 gcc_assert (single_exit (loop));
2444 entry = loop_preheader_edge (loop);
2445 exit = single_dom_exit (loop);
2447 /* This rewrites the body in terms of new variables. This has already
2448 been done for oacc_kernels_p in pass_lower_omp/lower_omp (). */
2449 if (!oacc_kernels_p)
2451 eliminate_local_variables (entry, exit);
2452 /* In the old loop, move all variables non-local to the loop to a
2453 structure and back, and create separate decls for the variables used in
2454 loop. */
2455 separate_decls_in_region (entry, exit, reduction_list, &arg_struct,
2456 &new_arg_struct, &clsn_data);
2458 else
2460 arg_struct = NULL_TREE;
2461 new_arg_struct = NULL_TREE;
2462 clsn_data.load = NULL_TREE;
2463 clsn_data.load_bb = exit->dest;
2464 clsn_data.store = NULL_TREE;
2465 clsn_data.store_bb = NULL;
2468 /* Create the parallel constructs. */
2469 loc = UNKNOWN_LOCATION;
2470 cond_stmt = last_stmt (loop->header);
2471 if (cond_stmt)
2472 loc = gimple_location (cond_stmt);
2473 create_parallel_loop (loop, create_loop_fn (loc), arg_struct, new_arg_struct,
2474 n_threads, loc, oacc_kernels_p);
2475 if (reduction_list->elements () > 0)
2476 create_call_for_reduction (loop, reduction_list, &clsn_data);
2478 scev_reset ();
2480 /* Free loop bound estimations that could contain references to
2481 removed statements. */
2482 free_numbers_of_iterations_estimates (cfun);
2485 /* Returns true when LOOP contains vector phi nodes. */
2487 static bool
2488 loop_has_vector_phi_nodes (struct loop *loop ATTRIBUTE_UNUSED)
2490 unsigned i;
2491 basic_block *bbs = get_loop_body_in_dom_order (loop);
2492 gphi_iterator gsi;
2493 bool res = true;
2495 for (i = 0; i < loop->num_nodes; i++)
2496 for (gsi = gsi_start_phis (bbs[i]); !gsi_end_p (gsi); gsi_next (&gsi))
2497 if (TREE_CODE (TREE_TYPE (PHI_RESULT (gsi.phi ()))) == VECTOR_TYPE)
2498 goto end;
2500 res = false;
2501 end:
2502 free (bbs);
2503 return res;
2506 /* Create a reduction_info struct, initialize it with REDUC_STMT
2507 and PHI, insert it to the REDUCTION_LIST. */
2509 static void
2510 build_new_reduction (reduction_info_table_type *reduction_list,
2511 gimple *reduc_stmt, gphi *phi)
2513 reduction_info **slot;
2514 struct reduction_info *new_reduction;
2515 enum tree_code reduction_code;
2517 gcc_assert (reduc_stmt);
2519 if (gimple_code (reduc_stmt) == GIMPLE_PHI)
2521 tree op1 = PHI_ARG_DEF (reduc_stmt, 0);
2522 gimple *def1 = SSA_NAME_DEF_STMT (op1);
2523 reduction_code = gimple_assign_rhs_code (def1);
2525 else
2526 reduction_code = gimple_assign_rhs_code (reduc_stmt);
2527 /* Check for OpenMP supported reduction. */
2528 switch (reduction_code)
2530 case PLUS_EXPR:
2531 case MULT_EXPR:
2532 case MAX_EXPR:
2533 case MIN_EXPR:
2534 case BIT_IOR_EXPR:
2535 case BIT_XOR_EXPR:
2536 case BIT_AND_EXPR:
2537 case TRUTH_OR_EXPR:
2538 case TRUTH_XOR_EXPR:
2539 case TRUTH_AND_EXPR:
2540 break;
2541 default:
2542 return;
2545 if (dump_file && (dump_flags & TDF_DETAILS))
2547 fprintf (dump_file,
2548 "Detected reduction. reduction stmt is:\n");
2549 print_gimple_stmt (dump_file, reduc_stmt, 0);
2550 fprintf (dump_file, "\n");
2553 new_reduction = XCNEW (struct reduction_info);
2555 new_reduction->reduc_stmt = reduc_stmt;
2556 new_reduction->reduc_phi = phi;
2557 new_reduction->reduc_version = SSA_NAME_VERSION (gimple_phi_result (phi));
2558 new_reduction->reduction_code = reduction_code;
2559 slot = reduction_list->find_slot (new_reduction, INSERT);
2560 *slot = new_reduction;
2563 /* Callback for htab_traverse. Sets gimple_uid of reduc_phi stmts. */
2566 set_reduc_phi_uids (reduction_info **slot, void *data ATTRIBUTE_UNUSED)
2568 struct reduction_info *const red = *slot;
2569 gimple_set_uid (red->reduc_phi, red->reduc_version);
2570 return 1;
2573 /* Return true if the type of reduction performed by STMT is suitable
2574 for this pass. */
2576 static bool
2577 valid_reduction_p (gimple *stmt)
2579 /* Parallelization would reassociate the operation, which isn't
2580 allowed for in-order reductions. */
2581 stmt_vec_info stmt_info = vinfo_for_stmt (stmt);
2582 vect_reduction_type reduc_type = STMT_VINFO_REDUC_TYPE (stmt_info);
2583 return reduc_type != FOLD_LEFT_REDUCTION;
2586 /* Detect all reductions in the LOOP, insert them into REDUCTION_LIST. */
2588 static void
2589 gather_scalar_reductions (loop_p loop, reduction_info_table_type *reduction_list)
2591 gphi_iterator gsi;
2592 loop_vec_info simple_loop_info;
2593 auto_vec<gphi *, 4> double_reduc_phis;
2594 auto_vec<gimple *, 4> double_reduc_stmts;
2596 if (!stmt_vec_info_vec.exists ())
2597 init_stmt_vec_info_vec ();
2599 simple_loop_info = vect_analyze_loop_form (loop);
2600 if (simple_loop_info == NULL)
2601 goto gather_done;
2603 for (gsi = gsi_start_phis (loop->header); !gsi_end_p (gsi); gsi_next (&gsi))
2605 gphi *phi = gsi.phi ();
2606 affine_iv iv;
2607 tree res = PHI_RESULT (phi);
2608 bool double_reduc;
2610 if (virtual_operand_p (res))
2611 continue;
2613 if (simple_iv (loop, loop, res, &iv, true))
2614 continue;
2616 gimple *reduc_stmt
2617 = vect_force_simple_reduction (simple_loop_info, phi,
2618 &double_reduc, true);
2619 if (!reduc_stmt || !valid_reduction_p (reduc_stmt))
2620 continue;
2622 if (double_reduc)
2624 if (loop->inner->inner != NULL)
2625 continue;
2627 double_reduc_phis.safe_push (phi);
2628 double_reduc_stmts.safe_push (reduc_stmt);
2629 continue;
2632 build_new_reduction (reduction_list, reduc_stmt, phi);
2634 delete simple_loop_info;
2636 if (!double_reduc_phis.is_empty ())
2638 simple_loop_info = vect_analyze_loop_form (loop->inner);
2639 if (simple_loop_info)
2641 gphi *phi;
2642 unsigned int i;
2644 FOR_EACH_VEC_ELT (double_reduc_phis, i, phi)
2646 affine_iv iv;
2647 tree res = PHI_RESULT (phi);
2648 bool double_reduc;
2650 use_operand_p use_p;
2651 gimple *inner_stmt;
2652 bool single_use_p = single_imm_use (res, &use_p, &inner_stmt);
2653 gcc_assert (single_use_p);
2654 if (gimple_code (inner_stmt) != GIMPLE_PHI)
2655 continue;
2656 gphi *inner_phi = as_a <gphi *> (inner_stmt);
2657 if (simple_iv (loop->inner, loop->inner, PHI_RESULT (inner_phi),
2658 &iv, true))
2659 continue;
2661 gimple *inner_reduc_stmt
2662 = vect_force_simple_reduction (simple_loop_info, inner_phi,
2663 &double_reduc, true);
2664 gcc_assert (!double_reduc);
2665 if (inner_reduc_stmt == NULL
2666 || !valid_reduction_p (inner_reduc_stmt))
2667 continue;
2669 build_new_reduction (reduction_list, double_reduc_stmts[i], phi);
2671 delete simple_loop_info;
2675 gather_done:
2676 /* Release the claim on gimple_uid. */
2677 free_stmt_vec_info_vec ();
2679 if (reduction_list->elements () == 0)
2680 return;
2682 /* As gimple_uid is used by the vectorizer in between vect_analyze_loop_form
2683 and free_stmt_vec_info_vec, we can set gimple_uid of reduc_phi stmts only
2684 now. */
2685 basic_block bb;
2686 FOR_EACH_BB_FN (bb, cfun)
2687 for (gsi = gsi_start_phis (bb); !gsi_end_p (gsi); gsi_next (&gsi))
2688 gimple_set_uid (gsi_stmt (gsi), (unsigned int)-1);
2689 reduction_list->traverse <void *, set_reduc_phi_uids> (NULL);
2692 /* Try to initialize NITER for code generation part. */
2694 static bool
2695 try_get_loop_niter (loop_p loop, struct tree_niter_desc *niter)
2697 edge exit = single_dom_exit (loop);
2699 gcc_assert (exit);
2701 /* We need to know # of iterations, and there should be no uses of values
2702 defined inside loop outside of it, unless the values are invariants of
2703 the loop. */
2704 if (!number_of_iterations_exit (loop, exit, niter, false))
2706 if (dump_file && (dump_flags & TDF_DETAILS))
2707 fprintf (dump_file, " FAILED: number of iterations not known\n");
2708 return false;
2711 return true;
2714 /* Return the default def of the first function argument. */
2716 static tree
2717 get_omp_data_i_param (void)
2719 tree decl = DECL_ARGUMENTS (cfun->decl);
2720 gcc_assert (DECL_CHAIN (decl) == NULL_TREE);
2721 return ssa_default_def (cfun, decl);
2724 /* For PHI in loop header of LOOP, look for pattern:
2726 <bb preheader>
2727 .omp_data_i = &.omp_data_arr;
2728 addr = .omp_data_i->sum;
2729 sum_a = *addr;
2731 <bb header>:
2732 sum_b = PHI <sum_a (preheader), sum_c (latch)>
2734 and return addr. Otherwise, return NULL_TREE. */
2736 static tree
2737 find_reduc_addr (struct loop *loop, gphi *phi)
2739 edge e = loop_preheader_edge (loop);
2740 tree arg = PHI_ARG_DEF_FROM_EDGE (phi, e);
2741 gimple *stmt = SSA_NAME_DEF_STMT (arg);
2742 if (!gimple_assign_single_p (stmt))
2743 return NULL_TREE;
2744 tree memref = gimple_assign_rhs1 (stmt);
2745 if (TREE_CODE (memref) != MEM_REF)
2746 return NULL_TREE;
2747 tree addr = TREE_OPERAND (memref, 0);
2749 gimple *stmt2 = SSA_NAME_DEF_STMT (addr);
2750 if (!gimple_assign_single_p (stmt2))
2751 return NULL_TREE;
2752 tree compref = gimple_assign_rhs1 (stmt2);
2753 if (TREE_CODE (compref) != COMPONENT_REF)
2754 return NULL_TREE;
2755 tree addr2 = TREE_OPERAND (compref, 0);
2756 if (TREE_CODE (addr2) != MEM_REF)
2757 return NULL_TREE;
2758 addr2 = TREE_OPERAND (addr2, 0);
2759 if (TREE_CODE (addr2) != SSA_NAME
2760 || addr2 != get_omp_data_i_param ())
2761 return NULL_TREE;
2763 return addr;
2766 /* Try to initialize REDUCTION_LIST for code generation part.
2767 REDUCTION_LIST describes the reductions. */
2769 static bool
2770 try_create_reduction_list (loop_p loop,
2771 reduction_info_table_type *reduction_list,
2772 bool oacc_kernels_p)
2774 edge exit = single_dom_exit (loop);
2775 gphi_iterator gsi;
2777 gcc_assert (exit);
2779 /* Try to get rid of exit phis. */
2780 final_value_replacement_loop (loop);
2782 gather_scalar_reductions (loop, reduction_list);
2785 for (gsi = gsi_start_phis (exit->dest); !gsi_end_p (gsi); gsi_next (&gsi))
2787 gphi *phi = gsi.phi ();
2788 struct reduction_info *red;
2789 imm_use_iterator imm_iter;
2790 use_operand_p use_p;
2791 gimple *reduc_phi;
2792 tree val = PHI_ARG_DEF_FROM_EDGE (phi, exit);
2794 if (!virtual_operand_p (val))
2796 if (dump_file && (dump_flags & TDF_DETAILS))
2798 fprintf (dump_file, "phi is ");
2799 print_gimple_stmt (dump_file, phi, 0);
2800 fprintf (dump_file, "arg of phi to exit: value ");
2801 print_generic_expr (dump_file, val);
2802 fprintf (dump_file, " used outside loop\n");
2803 fprintf (dump_file,
2804 " checking if it is part of reduction pattern:\n");
2806 if (reduction_list->elements () == 0)
2808 if (dump_file && (dump_flags & TDF_DETAILS))
2809 fprintf (dump_file,
2810 " FAILED: it is not a part of reduction.\n");
2811 return false;
2813 reduc_phi = NULL;
2814 FOR_EACH_IMM_USE_FAST (use_p, imm_iter, val)
2816 if (!gimple_debug_bind_p (USE_STMT (use_p))
2817 && flow_bb_inside_loop_p (loop, gimple_bb (USE_STMT (use_p))))
2819 reduc_phi = USE_STMT (use_p);
2820 break;
2823 red = reduction_phi (reduction_list, reduc_phi);
2824 if (red == NULL)
2826 if (dump_file && (dump_flags & TDF_DETAILS))
2827 fprintf (dump_file,
2828 " FAILED: it is not a part of reduction.\n");
2829 return false;
2831 if (red->keep_res != NULL)
2833 if (dump_file && (dump_flags & TDF_DETAILS))
2834 fprintf (dump_file,
2835 " FAILED: reduction has multiple exit phis.\n");
2836 return false;
2838 red->keep_res = phi;
2839 if (dump_file && (dump_flags & TDF_DETAILS))
2841 fprintf (dump_file, "reduction phi is ");
2842 print_gimple_stmt (dump_file, red->reduc_phi, 0);
2843 fprintf (dump_file, "reduction stmt is ");
2844 print_gimple_stmt (dump_file, red->reduc_stmt, 0);
2849 /* The iterations of the loop may communicate only through bivs whose
2850 iteration space can be distributed efficiently. */
2851 for (gsi = gsi_start_phis (loop->header); !gsi_end_p (gsi); gsi_next (&gsi))
2853 gphi *phi = gsi.phi ();
2854 tree def = PHI_RESULT (phi);
2855 affine_iv iv;
2857 if (!virtual_operand_p (def) && !simple_iv (loop, loop, def, &iv, true))
2859 struct reduction_info *red;
2861 red = reduction_phi (reduction_list, phi);
2862 if (red == NULL)
2864 if (dump_file && (dump_flags & TDF_DETAILS))
2865 fprintf (dump_file,
2866 " FAILED: scalar dependency between iterations\n");
2867 return false;
2872 if (oacc_kernels_p)
2874 for (gsi = gsi_start_phis (loop->header); !gsi_end_p (gsi);
2875 gsi_next (&gsi))
2877 gphi *phi = gsi.phi ();
2878 tree def = PHI_RESULT (phi);
2879 affine_iv iv;
2881 if (!virtual_operand_p (def)
2882 && !simple_iv (loop, loop, def, &iv, true))
2884 tree addr = find_reduc_addr (loop, phi);
2885 if (addr == NULL_TREE)
2886 return false;
2887 struct reduction_info *red = reduction_phi (reduction_list, phi);
2888 red->reduc_addr = addr;
2893 return true;
2896 /* Return true if LOOP contains phis with ADDR_EXPR in args. */
2898 static bool
2899 loop_has_phi_with_address_arg (struct loop *loop)
2901 basic_block *bbs = get_loop_body (loop);
2902 bool res = false;
2904 unsigned i, j;
2905 gphi_iterator gsi;
2906 for (i = 0; i < loop->num_nodes; i++)
2907 for (gsi = gsi_start_phis (bbs[i]); !gsi_end_p (gsi); gsi_next (&gsi))
2909 gphi *phi = gsi.phi ();
2910 for (j = 0; j < gimple_phi_num_args (phi); j++)
2912 tree arg = gimple_phi_arg_def (phi, j);
2913 if (TREE_CODE (arg) == ADDR_EXPR)
2915 /* This should be handled by eliminate_local_variables, but that
2916 function currently ignores phis. */
2917 res = true;
2918 goto end;
2922 end:
2923 free (bbs);
2925 return res;
2928 /* Return true if memory ref REF (corresponding to the stmt at GSI in
2929 REGIONS_BB[I]) conflicts with the statements in REGIONS_BB[I] after gsi,
2930 or the statements in REGIONS_BB[I + n]. REF_IS_STORE indicates if REF is a
2931 store. Ignore conflicts with SKIP_STMT. */
2933 static bool
2934 ref_conflicts_with_region (gimple_stmt_iterator gsi, ao_ref *ref,
2935 bool ref_is_store, vec<basic_block> region_bbs,
2936 unsigned int i, gimple *skip_stmt)
2938 basic_block bb = region_bbs[i];
2939 gsi_next (&gsi);
2941 while (true)
2943 for (; !gsi_end_p (gsi);
2944 gsi_next (&gsi))
2946 gimple *stmt = gsi_stmt (gsi);
2947 if (stmt == skip_stmt)
2949 if (dump_file)
2951 fprintf (dump_file, "skipping reduction store: ");
2952 print_gimple_stmt (dump_file, stmt, 0);
2954 continue;
2957 if (!gimple_vdef (stmt)
2958 && !gimple_vuse (stmt))
2959 continue;
2961 if (gimple_code (stmt) == GIMPLE_RETURN)
2962 continue;
2964 if (ref_is_store)
2966 if (ref_maybe_used_by_stmt_p (stmt, ref))
2968 if (dump_file)
2970 fprintf (dump_file, "Stmt ");
2971 print_gimple_stmt (dump_file, stmt, 0);
2973 return true;
2976 else
2978 if (stmt_may_clobber_ref_p_1 (stmt, ref))
2980 if (dump_file)
2982 fprintf (dump_file, "Stmt ");
2983 print_gimple_stmt (dump_file, stmt, 0);
2985 return true;
2989 i++;
2990 if (i == region_bbs.length ())
2991 break;
2992 bb = region_bbs[i];
2993 gsi = gsi_start_bb (bb);
2996 return false;
2999 /* Return true if the bbs in REGION_BBS but not in in_loop_bbs can be executed
3000 in parallel with REGION_BBS containing the loop. Return the stores of
3001 reduction results in REDUCTION_STORES. */
3003 static bool
3004 oacc_entry_exit_ok_1 (bitmap in_loop_bbs, vec<basic_block> region_bbs,
3005 reduction_info_table_type *reduction_list,
3006 bitmap reduction_stores)
3008 tree omp_data_i = get_omp_data_i_param ();
3010 unsigned i;
3011 basic_block bb;
3012 FOR_EACH_VEC_ELT (region_bbs, i, bb)
3014 if (bitmap_bit_p (in_loop_bbs, bb->index))
3015 continue;
3017 gimple_stmt_iterator gsi;
3018 for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi);
3019 gsi_next (&gsi))
3021 gimple *stmt = gsi_stmt (gsi);
3022 gimple *skip_stmt = NULL;
3024 if (is_gimple_debug (stmt)
3025 || gimple_code (stmt) == GIMPLE_COND)
3026 continue;
3028 ao_ref ref;
3029 bool ref_is_store = false;
3030 if (gimple_assign_load_p (stmt))
3032 tree rhs = gimple_assign_rhs1 (stmt);
3033 tree base = get_base_address (rhs);
3034 if (TREE_CODE (base) == MEM_REF
3035 && operand_equal_p (TREE_OPERAND (base, 0), omp_data_i, 0))
3036 continue;
3038 tree lhs = gimple_assign_lhs (stmt);
3039 if (TREE_CODE (lhs) == SSA_NAME
3040 && has_single_use (lhs))
3042 use_operand_p use_p;
3043 gimple *use_stmt;
3044 single_imm_use (lhs, &use_p, &use_stmt);
3045 if (gimple_code (use_stmt) == GIMPLE_PHI)
3047 struct reduction_info *red;
3048 red = reduction_phi (reduction_list, use_stmt);
3049 tree val = PHI_RESULT (red->keep_res);
3050 if (has_single_use (val))
3052 single_imm_use (val, &use_p, &use_stmt);
3053 if (gimple_store_p (use_stmt))
3055 unsigned int id
3056 = SSA_NAME_VERSION (gimple_vdef (use_stmt));
3057 bitmap_set_bit (reduction_stores, id);
3058 skip_stmt = use_stmt;
3059 if (dump_file)
3061 fprintf (dump_file, "found reduction load: ");
3062 print_gimple_stmt (dump_file, stmt, 0);
3069 ao_ref_init (&ref, rhs);
3071 else if (gimple_store_p (stmt))
3073 ao_ref_init (&ref, gimple_assign_lhs (stmt));
3074 ref_is_store = true;
3076 else if (gimple_code (stmt) == GIMPLE_OMP_RETURN)
3077 continue;
3078 else if (!gimple_has_side_effects (stmt)
3079 && !gimple_could_trap_p (stmt)
3080 && !stmt_could_throw_p (stmt)
3081 && !gimple_vdef (stmt)
3082 && !gimple_vuse (stmt))
3083 continue;
3084 else if (gimple_call_internal_p (stmt, IFN_GOACC_DIM_POS))
3085 continue;
3086 else if (gimple_code (stmt) == GIMPLE_RETURN)
3087 continue;
3088 else
3090 if (dump_file)
3092 fprintf (dump_file, "Unhandled stmt in entry/exit: ");
3093 print_gimple_stmt (dump_file, stmt, 0);
3095 return false;
3098 if (ref_conflicts_with_region (gsi, &ref, ref_is_store, region_bbs,
3099 i, skip_stmt))
3101 if (dump_file)
3103 fprintf (dump_file, "conflicts with entry/exit stmt: ");
3104 print_gimple_stmt (dump_file, stmt, 0);
3106 return false;
3111 return true;
3114 /* Find stores inside REGION_BBS and outside IN_LOOP_BBS, and guard them with
3115 gang_pos == 0, except when the stores are REDUCTION_STORES. Return true
3116 if any changes were made. */
3118 static bool
3119 oacc_entry_exit_single_gang (bitmap in_loop_bbs, vec<basic_block> region_bbs,
3120 bitmap reduction_stores)
3122 tree gang_pos = NULL_TREE;
3123 bool changed = false;
3125 unsigned i;
3126 basic_block bb;
3127 FOR_EACH_VEC_ELT (region_bbs, i, bb)
3129 if (bitmap_bit_p (in_loop_bbs, bb->index))
3130 continue;
3132 gimple_stmt_iterator gsi;
3133 for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi);)
3135 gimple *stmt = gsi_stmt (gsi);
3137 if (!gimple_store_p (stmt))
3139 /* Update gsi to point to next stmt. */
3140 gsi_next (&gsi);
3141 continue;
3144 if (bitmap_bit_p (reduction_stores,
3145 SSA_NAME_VERSION (gimple_vdef (stmt))))
3147 if (dump_file)
3149 fprintf (dump_file,
3150 "skipped reduction store for single-gang"
3151 " neutering: ");
3152 print_gimple_stmt (dump_file, stmt, 0);
3155 /* Update gsi to point to next stmt. */
3156 gsi_next (&gsi);
3157 continue;
3160 changed = true;
3162 if (gang_pos == NULL_TREE)
3164 tree arg = build_int_cst (integer_type_node, GOMP_DIM_GANG);
3165 gcall *gang_single
3166 = gimple_build_call_internal (IFN_GOACC_DIM_POS, 1, arg);
3167 gang_pos = make_ssa_name (integer_type_node);
3168 gimple_call_set_lhs (gang_single, gang_pos);
3169 gimple_stmt_iterator start
3170 = gsi_start_bb (single_succ (ENTRY_BLOCK_PTR_FOR_FN (cfun)));
3171 tree vuse = ssa_default_def (cfun, gimple_vop (cfun));
3172 gimple_set_vuse (gang_single, vuse);
3173 gsi_insert_before (&start, gang_single, GSI_SAME_STMT);
3176 if (dump_file)
3178 fprintf (dump_file,
3179 "found store that needs single-gang neutering: ");
3180 print_gimple_stmt (dump_file, stmt, 0);
3184 /* Split block before store. */
3185 gimple_stmt_iterator gsi2 = gsi;
3186 gsi_prev (&gsi2);
3187 edge e;
3188 if (gsi_end_p (gsi2))
3190 e = split_block_after_labels (bb);
3191 gsi2 = gsi_last_bb (bb);
3193 else
3194 e = split_block (bb, gsi_stmt (gsi2));
3195 basic_block bb2 = e->dest;
3197 /* Split block after store. */
3198 gimple_stmt_iterator gsi3 = gsi_start_bb (bb2);
3199 edge e2 = split_block (bb2, gsi_stmt (gsi3));
3200 basic_block bb3 = e2->dest;
3202 gimple *cond
3203 = gimple_build_cond (EQ_EXPR, gang_pos, integer_zero_node,
3204 NULL_TREE, NULL_TREE);
3205 gsi_insert_after (&gsi2, cond, GSI_NEW_STMT);
3207 edge e3 = make_edge (bb, bb3, EDGE_FALSE_VALUE);
3208 /* FIXME: What is the probability? */
3209 e3->probability = profile_probability::guessed_never ();
3210 e->flags = EDGE_TRUE_VALUE;
3212 tree vdef = gimple_vdef (stmt);
3213 tree vuse = gimple_vuse (stmt);
3215 tree phi_res = copy_ssa_name (vdef);
3216 gphi *new_phi = create_phi_node (phi_res, bb3);
3217 replace_uses_by (vdef, phi_res);
3218 add_phi_arg (new_phi, vuse, e3, UNKNOWN_LOCATION);
3219 add_phi_arg (new_phi, vdef, e2, UNKNOWN_LOCATION);
3221 /* Update gsi to point to next stmt. */
3222 bb = bb3;
3223 gsi = gsi_start_bb (bb);
3228 return changed;
3231 /* Return true if the statements before and after the LOOP can be executed in
3232 parallel with the function containing the loop. Resolve conflicting stores
3233 outside LOOP by guarding them such that only a single gang executes them. */
3235 static bool
3236 oacc_entry_exit_ok (struct loop *loop,
3237 reduction_info_table_type *reduction_list)
3239 basic_block *loop_bbs = get_loop_body_in_dom_order (loop);
3240 vec<basic_block> region_bbs
3241 = get_all_dominated_blocks (CDI_DOMINATORS, ENTRY_BLOCK_PTR_FOR_FN (cfun));
3243 bitmap in_loop_bbs = BITMAP_ALLOC (NULL);
3244 bitmap_clear (in_loop_bbs);
3245 for (unsigned int i = 0; i < loop->num_nodes; i++)
3246 bitmap_set_bit (in_loop_bbs, loop_bbs[i]->index);
3248 bitmap reduction_stores = BITMAP_ALLOC (NULL);
3249 bool res = oacc_entry_exit_ok_1 (in_loop_bbs, region_bbs, reduction_list,
3250 reduction_stores);
3252 if (res)
3254 bool changed = oacc_entry_exit_single_gang (in_loop_bbs, region_bbs,
3255 reduction_stores);
3256 if (changed)
3258 free_dominance_info (CDI_DOMINATORS);
3259 calculate_dominance_info (CDI_DOMINATORS);
3263 region_bbs.release ();
3264 free (loop_bbs);
3266 BITMAP_FREE (in_loop_bbs);
3267 BITMAP_FREE (reduction_stores);
3269 return res;
3272 /* Detect parallel loops and generate parallel code using libgomp
3273 primitives. Returns true if some loop was parallelized, false
3274 otherwise. */
3276 static bool
3277 parallelize_loops (bool oacc_kernels_p)
3279 unsigned n_threads;
3280 bool changed = false;
3281 struct loop *loop;
3282 struct loop *skip_loop = NULL;
3283 struct tree_niter_desc niter_desc;
3284 struct obstack parloop_obstack;
3285 HOST_WIDE_INT estimated;
3286 source_location loop_loc;
3288 /* Do not parallelize loops in the functions created by parallelization. */
3289 if (!oacc_kernels_p
3290 && parallelized_function_p (cfun->decl))
3291 return false;
3293 /* Do not parallelize loops in offloaded functions. */
3294 if (!oacc_kernels_p
3295 && oacc_get_fn_attrib (cfun->decl) != NULL)
3296 return false;
3298 if (cfun->has_nonlocal_label)
3299 return false;
3301 /* For OpenACC kernels, n_threads will be determined later; otherwise, it's
3302 the argument to -ftree-parallelize-loops. */
3303 if (oacc_kernels_p)
3304 n_threads = 0;
3305 else
3306 n_threads = flag_tree_parallelize_loops;
3308 gcc_obstack_init (&parloop_obstack);
3309 reduction_info_table_type reduction_list (10);
3311 calculate_dominance_info (CDI_DOMINATORS);
3313 FOR_EACH_LOOP (loop, 0)
3315 if (loop == skip_loop)
3317 if (!loop->in_oacc_kernels_region
3318 && dump_file && (dump_flags & TDF_DETAILS))
3319 fprintf (dump_file,
3320 "Skipping loop %d as inner loop of parallelized loop\n",
3321 loop->num);
3323 skip_loop = loop->inner;
3324 continue;
3326 else
3327 skip_loop = NULL;
3329 reduction_list.empty ();
3331 if (oacc_kernels_p)
3333 if (!loop->in_oacc_kernels_region)
3334 continue;
3336 /* Don't try to parallelize inner loops in an oacc kernels region. */
3337 if (loop->inner)
3338 skip_loop = loop->inner;
3340 if (dump_file && (dump_flags & TDF_DETAILS))
3341 fprintf (dump_file,
3342 "Trying loop %d with header bb %d in oacc kernels"
3343 " region\n", loop->num, loop->header->index);
3346 if (dump_file && (dump_flags & TDF_DETAILS))
3348 fprintf (dump_file, "Trying loop %d as candidate\n",loop->num);
3349 if (loop->inner)
3350 fprintf (dump_file, "loop %d is not innermost\n",loop->num);
3351 else
3352 fprintf (dump_file, "loop %d is innermost\n",loop->num);
3355 if (!single_dom_exit (loop))
3358 if (dump_file && (dump_flags & TDF_DETAILS))
3359 fprintf (dump_file, "loop is !single_dom_exit\n");
3361 continue;
3364 if (/* And of course, the loop must be parallelizable. */
3365 !can_duplicate_loop_p (loop)
3366 || loop_has_blocks_with_irreducible_flag (loop)
3367 || (loop_preheader_edge (loop)->src->flags & BB_IRREDUCIBLE_LOOP)
3368 /* FIXME: the check for vector phi nodes could be removed. */
3369 || loop_has_vector_phi_nodes (loop))
3370 continue;
3372 estimated = estimated_loop_iterations_int (loop);
3373 if (estimated == -1)
3374 estimated = get_likely_max_loop_iterations_int (loop);
3375 /* FIXME: Bypass this check as graphite doesn't update the
3376 count and frequency correctly now. */
3377 if (!flag_loop_parallelize_all
3378 && !oacc_kernels_p
3379 && ((estimated != -1
3380 && (estimated
3381 < ((HOST_WIDE_INT) n_threads
3382 * (loop->inner ? 2 : MIN_PER_THREAD) - 1)))
3383 /* Do not bother with loops in cold areas. */
3384 || optimize_loop_nest_for_size_p (loop)))
3385 continue;
3387 if (!try_get_loop_niter (loop, &niter_desc))
3388 continue;
3390 if (!try_create_reduction_list (loop, &reduction_list, oacc_kernels_p))
3391 continue;
3393 if (loop_has_phi_with_address_arg (loop))
3394 continue;
3396 if (!loop->can_be_parallel
3397 && !loop_parallel_p (loop, &parloop_obstack))
3398 continue;
3400 if (oacc_kernels_p
3401 && !oacc_entry_exit_ok (loop, &reduction_list))
3403 if (dump_file)
3404 fprintf (dump_file, "entry/exit not ok: FAILED\n");
3405 continue;
3408 changed = true;
3409 skip_loop = loop->inner;
3411 loop_loc = find_loop_location (loop);
3412 if (loop->inner)
3413 dump_printf_loc (MSG_OPTIMIZED_LOCATIONS, loop_loc,
3414 "parallelizing outer loop %d\n", loop->num);
3415 else
3416 dump_printf_loc (MSG_OPTIMIZED_LOCATIONS, loop_loc,
3417 "parallelizing inner loop %d\n", loop->num);
3419 gen_parallel_loop (loop, &reduction_list,
3420 n_threads, &niter_desc, oacc_kernels_p);
3423 obstack_free (&parloop_obstack, NULL);
3425 /* Parallelization will cause new function calls to be inserted through
3426 which local variables will escape. Reset the points-to solution
3427 for ESCAPED. */
3428 if (changed)
3429 pt_solution_reset (&cfun->gimple_df->escaped);
3431 return changed;
3434 /* Parallelization. */
3436 namespace {
3438 const pass_data pass_data_parallelize_loops =
3440 GIMPLE_PASS, /* type */
3441 "parloops", /* name */
3442 OPTGROUP_LOOP, /* optinfo_flags */
3443 TV_TREE_PARALLELIZE_LOOPS, /* tv_id */
3444 ( PROP_cfg | PROP_ssa ), /* properties_required */
3445 0, /* properties_provided */
3446 0, /* properties_destroyed */
3447 0, /* todo_flags_start */
3448 0, /* todo_flags_finish */
3451 class pass_parallelize_loops : public gimple_opt_pass
3453 public:
3454 pass_parallelize_loops (gcc::context *ctxt)
3455 : gimple_opt_pass (pass_data_parallelize_loops, ctxt),
3456 oacc_kernels_p (false)
3459 /* opt_pass methods: */
3460 virtual bool gate (function *)
3462 if (oacc_kernels_p)
3463 return flag_openacc;
3464 else
3465 return flag_tree_parallelize_loops > 1;
3467 virtual unsigned int execute (function *);
3468 opt_pass * clone () { return new pass_parallelize_loops (m_ctxt); }
3469 void set_pass_param (unsigned int n, bool param)
3471 gcc_assert (n == 0);
3472 oacc_kernels_p = param;
3475 private:
3476 bool oacc_kernels_p;
3477 }; // class pass_parallelize_loops
3479 unsigned
3480 pass_parallelize_loops::execute (function *fun)
3482 tree nthreads = builtin_decl_explicit (BUILT_IN_OMP_GET_NUM_THREADS);
3483 if (nthreads == NULL_TREE)
3484 return 0;
3486 bool in_loop_pipeline = scev_initialized_p ();
3487 if (!in_loop_pipeline)
3488 loop_optimizer_init (LOOPS_NORMAL
3489 | LOOPS_HAVE_RECORDED_EXITS);
3491 if (number_of_loops (fun) <= 1)
3492 return 0;
3494 if (!in_loop_pipeline)
3496 rewrite_into_loop_closed_ssa (NULL, TODO_update_ssa);
3497 scev_initialize ();
3500 unsigned int todo = 0;
3501 if (parallelize_loops (oacc_kernels_p))
3503 fun->curr_properties &= ~(PROP_gimple_eomp);
3505 checking_verify_loop_structure ();
3507 todo |= TODO_update_ssa;
3510 if (!in_loop_pipeline)
3512 scev_finalize ();
3513 loop_optimizer_finalize ();
3516 return todo;
3519 } // anon namespace
3521 gimple_opt_pass *
3522 make_pass_parallelize_loops (gcc::context *ctxt)
3524 return new pass_parallelize_loops (ctxt);