Enable dumping of alias graphs.
[official-gcc/Ramakrishna.git] / gcc / tree-scalar-evolution.c
blobfcd3382836ee9cca3ed54db3ba25f3b8a7c81fce
1 /* Scalar evolution detector.
2 Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009
3 Free Software Foundation, Inc.
4 Contributed by Sebastian Pop <s.pop@laposte.net>
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 /*
23 Description:
25 This pass analyzes the evolution of scalar variables in loop
26 structures. The algorithm is based on the SSA representation,
27 and on the loop hierarchy tree. This algorithm is not based on
28 the notion of versions of a variable, as it was the case for the
29 previous implementations of the scalar evolution algorithm, but
30 it assumes that each defined name is unique.
32 The notation used in this file is called "chains of recurrences",
33 and has been proposed by Eugene Zima, Robert Van Engelen, and
34 others for describing induction variables in programs. For example
35 "b -> {0, +, 2}_1" means that the scalar variable "b" is equal to 0
36 when entering in the loop_1 and has a step 2 in this loop, in other
37 words "for (b = 0; b < N; b+=2);". Note that the coefficients of
38 this chain of recurrence (or chrec [shrek]) can contain the name of
39 other variables, in which case they are called parametric chrecs.
40 For example, "b -> {a, +, 2}_1" means that the initial value of "b"
41 is the value of "a". In most of the cases these parametric chrecs
42 are fully instantiated before their use because symbolic names can
43 hide some difficult cases such as self-references described later
44 (see the Fibonacci example).
46 A short sketch of the algorithm is:
48 Given a scalar variable to be analyzed, follow the SSA edge to
49 its definition:
51 - When the definition is a GIMPLE_ASSIGN: if the right hand side
52 (RHS) of the definition cannot be statically analyzed, the answer
53 of the analyzer is: "don't know".
54 Otherwise, for all the variables that are not yet analyzed in the
55 RHS, try to determine their evolution, and finally try to
56 evaluate the operation of the RHS that gives the evolution
57 function of the analyzed variable.
59 - When the definition is a condition-phi-node: determine the
60 evolution function for all the branches of the phi node, and
61 finally merge these evolutions (see chrec_merge).
63 - When the definition is a loop-phi-node: determine its initial
64 condition, that is the SSA edge defined in an outer loop, and
65 keep it symbolic. Then determine the SSA edges that are defined
66 in the body of the loop. Follow the inner edges until ending on
67 another loop-phi-node of the same analyzed loop. If the reached
68 loop-phi-node is not the starting loop-phi-node, then we keep
69 this definition under a symbolic form. If the reached
70 loop-phi-node is the same as the starting one, then we compute a
71 symbolic stride on the return path. The result is then the
72 symbolic chrec {initial_condition, +, symbolic_stride}_loop.
74 Examples:
76 Example 1: Illustration of the basic algorithm.
78 | a = 3
79 | loop_1
80 | b = phi (a, c)
81 | c = b + 1
82 | if (c > 10) exit_loop
83 | endloop
85 Suppose that we want to know the number of iterations of the
86 loop_1. The exit_loop is controlled by a COND_EXPR (c > 10). We
87 ask the scalar evolution analyzer two questions: what's the
88 scalar evolution (scev) of "c", and what's the scev of "10". For
89 "10" the answer is "10" since it is a scalar constant. For the
90 scalar variable "c", it follows the SSA edge to its definition,
91 "c = b + 1", and then asks again what's the scev of "b".
92 Following the SSA edge, we end on a loop-phi-node "b = phi (a,
93 c)", where the initial condition is "a", and the inner loop edge
94 is "c". The initial condition is kept under a symbolic form (it
95 may be the case that the copy constant propagation has done its
96 work and we end with the constant "3" as one of the edges of the
97 loop-phi-node). The update edge is followed to the end of the
98 loop, and until reaching again the starting loop-phi-node: b -> c
99 -> b. At this point we have drawn a path from "b" to "b" from
100 which we compute the stride in the loop: in this example it is
101 "+1". The resulting scev for "b" is "b -> {a, +, 1}_1". Now
102 that the scev for "b" is known, it is possible to compute the
103 scev for "c", that is "c -> {a + 1, +, 1}_1". In order to
104 determine the number of iterations in the loop_1, we have to
105 instantiate_parameters (loop_1, {a + 1, +, 1}_1), that gives after some
106 more analysis the scev {4, +, 1}_1, or in other words, this is
107 the function "f (x) = x + 4", where x is the iteration count of
108 the loop_1. Now we have to solve the inequality "x + 4 > 10",
109 and take the smallest iteration number for which the loop is
110 exited: x = 7. This loop runs from x = 0 to x = 7, and in total
111 there are 8 iterations. In terms of loop normalization, we have
112 created a variable that is implicitly defined, "x" or just "_1",
113 and all the other analyzed scalars of the loop are defined in
114 function of this variable:
116 a -> 3
117 b -> {3, +, 1}_1
118 c -> {4, +, 1}_1
120 or in terms of a C program:
122 | a = 3
123 | for (x = 0; x <= 7; x++)
125 | b = x + 3
126 | c = x + 4
129 Example 2a: Illustration of the algorithm on nested loops.
131 | loop_1
132 | a = phi (1, b)
133 | c = a + 2
134 | loop_2 10 times
135 | b = phi (c, d)
136 | d = b + 3
137 | endloop
138 | endloop
140 For analyzing the scalar evolution of "a", the algorithm follows
141 the SSA edge into the loop's body: "a -> b". "b" is an inner
142 loop-phi-node, and its analysis as in Example 1, gives:
144 b -> {c, +, 3}_2
145 d -> {c + 3, +, 3}_2
147 Following the SSA edge for the initial condition, we end on "c = a
148 + 2", and then on the starting loop-phi-node "a". From this point,
149 the loop stride is computed: back on "c = a + 2" we get a "+2" in
150 the loop_1, then on the loop-phi-node "b" we compute the overall
151 effect of the inner loop that is "b = c + 30", and we get a "+30"
152 in the loop_1. That means that the overall stride in loop_1 is
153 equal to "+32", and the result is:
155 a -> {1, +, 32}_1
156 c -> {3, +, 32}_1
158 Example 2b: Multivariate chains of recurrences.
160 | loop_1
161 | k = phi (0, k + 1)
162 | loop_2 4 times
163 | j = phi (0, j + 1)
164 | loop_3 4 times
165 | i = phi (0, i + 1)
166 | A[j + k] = ...
167 | endloop
168 | endloop
169 | endloop
171 Analyzing the access function of array A with
172 instantiate_parameters (loop_1, "j + k"), we obtain the
173 instantiation and the analysis of the scalar variables "j" and "k"
174 in loop_1. This leads to the scalar evolution {4, +, 1}_1: the end
175 value of loop_2 for "j" is 4, and the evolution of "k" in loop_1 is
176 {0, +, 1}_1. To obtain the evolution function in loop_3 and
177 instantiate the scalar variables up to loop_1, one has to use:
178 instantiate_scev (block_before_loop (loop_1), loop_3, "j + k").
179 The result of this call is {{0, +, 1}_1, +, 1}_2.
181 Example 3: Higher degree polynomials.
183 | loop_1
184 | a = phi (2, b)
185 | c = phi (5, d)
186 | b = a + 1
187 | d = c + a
188 | endloop
190 a -> {2, +, 1}_1
191 b -> {3, +, 1}_1
192 c -> {5, +, a}_1
193 d -> {5 + a, +, a}_1
195 instantiate_parameters (loop_1, {5, +, a}_1) -> {5, +, 2, +, 1}_1
196 instantiate_parameters (loop_1, {5 + a, +, a}_1) -> {7, +, 3, +, 1}_1
198 Example 4: Lucas, Fibonacci, or mixers in general.
200 | loop_1
201 | a = phi (1, b)
202 | c = phi (3, d)
203 | b = c
204 | d = c + a
205 | endloop
207 a -> (1, c)_1
208 c -> {3, +, a}_1
210 The syntax "(1, c)_1" stands for a PEELED_CHREC that has the
211 following semantics: during the first iteration of the loop_1, the
212 variable contains the value 1, and then it contains the value "c".
213 Note that this syntax is close to the syntax of the loop-phi-node:
214 "a -> (1, c)_1" vs. "a = phi (1, c)".
216 The symbolic chrec representation contains all the semantics of the
217 original code. What is more difficult is to use this information.
219 Example 5: Flip-flops, or exchangers.
221 | loop_1
222 | a = phi (1, b)
223 | c = phi (3, d)
224 | b = c
225 | d = a
226 | endloop
228 a -> (1, c)_1
229 c -> (3, a)_1
231 Based on these symbolic chrecs, it is possible to refine this
232 information into the more precise PERIODIC_CHRECs:
234 a -> |1, 3|_1
235 c -> |3, 1|_1
237 This transformation is not yet implemented.
239 Further readings:
241 You can find a more detailed description of the algorithm in:
242 http://icps.u-strasbg.fr/~pop/DEA_03_Pop.pdf
243 http://icps.u-strasbg.fr/~pop/DEA_03_Pop.ps.gz. But note that
244 this is a preliminary report and some of the details of the
245 algorithm have changed. I'm working on a research report that
246 updates the description of the algorithms to reflect the design
247 choices used in this implementation.
249 A set of slides show a high level overview of the algorithm and run
250 an example through the scalar evolution analyzer:
251 http://cri.ensmp.fr/~pop/gcc/mar04/slides.pdf
253 The slides that I have presented at the GCC Summit'04 are available
254 at: http://cri.ensmp.fr/~pop/gcc/20040604/gccsummit-lno-spop.pdf
257 #include "config.h"
258 #include "system.h"
259 #include "coretypes.h"
260 #include "tm.h"
261 #include "ggc.h"
262 #include "tree.h"
263 #include "real.h"
265 /* These RTL headers are needed for basic-block.h. */
266 #include "rtl.h"
267 #include "basic-block.h"
268 #include "diagnostic.h"
269 #include "tree-flow.h"
270 #include "tree-dump.h"
271 #include "timevar.h"
272 #include "cfgloop.h"
273 #include "tree-chrec.h"
274 #include "tree-scalar-evolution.h"
275 #include "tree-pass.h"
276 #include "flags.h"
277 #include "params.h"
279 static tree analyze_scalar_evolution_1 (struct loop *, tree, tree);
281 /* The cached information about an SSA name VAR, claiming that below
282 basic block INSTANTIATED_BELOW, the value of VAR can be expressed
283 as CHREC. */
285 struct GTY(()) scev_info_str {
286 basic_block instantiated_below;
287 tree var;
288 tree chrec;
291 /* Counters for the scev database. */
292 static unsigned nb_set_scev = 0;
293 static unsigned nb_get_scev = 0;
295 /* The following trees are unique elements. Thus the comparison of
296 another element to these elements should be done on the pointer to
297 these trees, and not on their value. */
299 /* The SSA_NAMEs that are not yet analyzed are qualified with NULL_TREE. */
300 tree chrec_not_analyzed_yet;
302 /* Reserved to the cases where the analyzer has detected an
303 undecidable property at compile time. */
304 tree chrec_dont_know;
306 /* When the analyzer has detected that a property will never
307 happen, then it qualifies it with chrec_known. */
308 tree chrec_known;
310 static GTY ((param_is (struct scev_info_str))) htab_t scalar_evolution_info;
313 /* Constructs a new SCEV_INFO_STR structure for VAR and INSTANTIATED_BELOW. */
315 static inline struct scev_info_str *
316 new_scev_info_str (basic_block instantiated_below, tree var)
318 struct scev_info_str *res;
320 res = GGC_NEW (struct scev_info_str);
321 res->var = var;
322 res->chrec = chrec_not_analyzed_yet;
323 res->instantiated_below = instantiated_below;
325 return res;
328 /* Computes a hash function for database element ELT. */
330 static hashval_t
331 hash_scev_info (const void *elt)
333 return SSA_NAME_VERSION (((const struct scev_info_str *) elt)->var);
336 /* Compares database elements E1 and E2. */
338 static int
339 eq_scev_info (const void *e1, const void *e2)
341 const struct scev_info_str *elt1 = (const struct scev_info_str *) e1;
342 const struct scev_info_str *elt2 = (const struct scev_info_str *) e2;
344 return (elt1->var == elt2->var
345 && elt1->instantiated_below == elt2->instantiated_below);
348 /* Deletes database element E. */
350 static void
351 del_scev_info (void *e)
353 ggc_free (e);
356 /* Get the scalar evolution of VAR for INSTANTIATED_BELOW basic block.
357 A first query on VAR returns chrec_not_analyzed_yet. */
359 static tree *
360 find_var_scev_info (basic_block instantiated_below, tree var)
362 struct scev_info_str *res;
363 struct scev_info_str tmp;
364 PTR *slot;
366 tmp.var = var;
367 tmp.instantiated_below = instantiated_below;
368 slot = htab_find_slot (scalar_evolution_info, &tmp, INSERT);
370 if (!*slot)
371 *slot = new_scev_info_str (instantiated_below, var);
372 res = (struct scev_info_str *) *slot;
374 return &res->chrec;
377 /* Return true when CHREC contains symbolic names defined in
378 LOOP_NB. */
380 bool
381 chrec_contains_symbols_defined_in_loop (const_tree chrec, unsigned loop_nb)
383 int i, n;
385 if (chrec == NULL_TREE)
386 return false;
388 if (is_gimple_min_invariant (chrec))
389 return false;
391 if (TREE_CODE (chrec) == VAR_DECL
392 || TREE_CODE (chrec) == PARM_DECL
393 || TREE_CODE (chrec) == FUNCTION_DECL
394 || TREE_CODE (chrec) == LABEL_DECL
395 || TREE_CODE (chrec) == RESULT_DECL
396 || TREE_CODE (chrec) == FIELD_DECL)
397 return true;
399 if (TREE_CODE (chrec) == SSA_NAME)
401 gimple def = SSA_NAME_DEF_STMT (chrec);
402 struct loop *def_loop = loop_containing_stmt (def);
403 struct loop *loop = get_loop (loop_nb);
405 if (def_loop == NULL)
406 return false;
408 if (loop == def_loop || flow_loop_nested_p (loop, def_loop))
409 return true;
411 return false;
414 n = TREE_OPERAND_LENGTH (chrec);
415 for (i = 0; i < n; i++)
416 if (chrec_contains_symbols_defined_in_loop (TREE_OPERAND (chrec, i),
417 loop_nb))
418 return true;
419 return false;
422 /* Return true when PHI is a loop-phi-node. */
424 static bool
425 loop_phi_node_p (gimple phi)
427 /* The implementation of this function is based on the following
428 property: "all the loop-phi-nodes of a loop are contained in the
429 loop's header basic block". */
431 return loop_containing_stmt (phi)->header == gimple_bb (phi);
434 /* Compute the scalar evolution for EVOLUTION_FN after crossing LOOP.
435 In general, in the case of multivariate evolutions we want to get
436 the evolution in different loops. LOOP specifies the level for
437 which to get the evolution.
439 Example:
441 | for (j = 0; j < 100; j++)
443 | for (k = 0; k < 100; k++)
445 | i = k + j; - Here the value of i is a function of j, k.
447 | ... = i - Here the value of i is a function of j.
449 | ... = i - Here the value of i is a scalar.
451 Example:
453 | i_0 = ...
454 | loop_1 10 times
455 | i_1 = phi (i_0, i_2)
456 | i_2 = i_1 + 2
457 | endloop
459 This loop has the same effect as:
460 LOOP_1 has the same effect as:
462 | i_1 = i_0 + 20
464 The overall effect of the loop, "i_0 + 20" in the previous example,
465 is obtained by passing in the parameters: LOOP = 1,
466 EVOLUTION_FN = {i_0, +, 2}_1.
469 tree
470 compute_overall_effect_of_inner_loop (struct loop *loop, tree evolution_fn)
472 bool val = false;
474 if (evolution_fn == chrec_dont_know)
475 return chrec_dont_know;
477 else if (TREE_CODE (evolution_fn) == POLYNOMIAL_CHREC)
479 struct loop *inner_loop = get_chrec_loop (evolution_fn);
481 if (inner_loop == loop
482 || flow_loop_nested_p (loop, inner_loop))
484 tree nb_iter = number_of_latch_executions (inner_loop);
486 if (nb_iter == chrec_dont_know)
487 return chrec_dont_know;
488 else
490 tree res;
492 /* evolution_fn is the evolution function in LOOP. Get
493 its value in the nb_iter-th iteration. */
494 res = chrec_apply (inner_loop->num, evolution_fn, nb_iter);
496 if (chrec_contains_symbols_defined_in_loop (res, loop->num))
497 res = instantiate_parameters (loop, res);
499 /* Continue the computation until ending on a parent of LOOP. */
500 return compute_overall_effect_of_inner_loop (loop, res);
503 else
504 return evolution_fn;
507 /* If the evolution function is an invariant, there is nothing to do. */
508 else if (no_evolution_in_loop_p (evolution_fn, loop->num, &val) && val)
509 return evolution_fn;
511 else
512 return chrec_dont_know;
515 /* Determine whether the CHREC is always positive/negative. If the expression
516 cannot be statically analyzed, return false, otherwise set the answer into
517 VALUE. */
519 bool
520 chrec_is_positive (tree chrec, bool *value)
522 bool value0, value1, value2;
523 tree end_value, nb_iter;
525 switch (TREE_CODE (chrec))
527 case POLYNOMIAL_CHREC:
528 if (!chrec_is_positive (CHREC_LEFT (chrec), &value0)
529 || !chrec_is_positive (CHREC_RIGHT (chrec), &value1))
530 return false;
532 /* FIXME -- overflows. */
533 if (value0 == value1)
535 *value = value0;
536 return true;
539 /* Otherwise the chrec is under the form: "{-197, +, 2}_1",
540 and the proof consists in showing that the sign never
541 changes during the execution of the loop, from 0 to
542 loop->nb_iterations. */
543 if (!evolution_function_is_affine_p (chrec))
544 return false;
546 nb_iter = number_of_latch_executions (get_chrec_loop (chrec));
547 if (chrec_contains_undetermined (nb_iter))
548 return false;
550 #if 0
551 /* TODO -- If the test is after the exit, we may decrease the number of
552 iterations by one. */
553 if (after_exit)
554 nb_iter = chrec_fold_minus (type, nb_iter, build_int_cst (type, 1));
555 #endif
557 end_value = chrec_apply (CHREC_VARIABLE (chrec), chrec, nb_iter);
559 if (!chrec_is_positive (end_value, &value2))
560 return false;
562 *value = value0;
563 return value0 == value1;
565 case INTEGER_CST:
566 *value = (tree_int_cst_sgn (chrec) == 1);
567 return true;
569 default:
570 return false;
574 /* Associate CHREC to SCALAR. */
576 static void
577 set_scalar_evolution (basic_block instantiated_below, tree scalar, tree chrec)
579 tree *scalar_info;
581 if (TREE_CODE (scalar) != SSA_NAME)
582 return;
584 scalar_info = find_var_scev_info (instantiated_below, scalar);
586 if (dump_file)
588 if (dump_flags & TDF_DETAILS)
590 fprintf (dump_file, "(set_scalar_evolution \n");
591 fprintf (dump_file, " instantiated_below = %d \n",
592 instantiated_below->index);
593 fprintf (dump_file, " (scalar = ");
594 print_generic_expr (dump_file, scalar, 0);
595 fprintf (dump_file, ")\n (scalar_evolution = ");
596 print_generic_expr (dump_file, chrec, 0);
597 fprintf (dump_file, "))\n");
599 if (dump_flags & TDF_STATS)
600 nb_set_scev++;
603 *scalar_info = chrec;
606 /* Retrieve the chrec associated to SCALAR instantiated below
607 INSTANTIATED_BELOW block. */
609 static tree
610 get_scalar_evolution (basic_block instantiated_below, tree scalar)
612 tree res;
614 if (dump_file)
616 if (dump_flags & TDF_DETAILS)
618 fprintf (dump_file, "(get_scalar_evolution \n");
619 fprintf (dump_file, " (scalar = ");
620 print_generic_expr (dump_file, scalar, 0);
621 fprintf (dump_file, ")\n");
623 if (dump_flags & TDF_STATS)
624 nb_get_scev++;
627 switch (TREE_CODE (scalar))
629 case SSA_NAME:
630 res = *find_var_scev_info (instantiated_below, scalar);
631 break;
633 case REAL_CST:
634 case FIXED_CST:
635 case INTEGER_CST:
636 res = scalar;
637 break;
639 default:
640 res = chrec_not_analyzed_yet;
641 break;
644 if (dump_file && (dump_flags & TDF_DETAILS))
646 fprintf (dump_file, " (scalar_evolution = ");
647 print_generic_expr (dump_file, res, 0);
648 fprintf (dump_file, "))\n");
651 return res;
654 /* Helper function for add_to_evolution. Returns the evolution
655 function for an assignment of the form "a = b + c", where "a" and
656 "b" are on the strongly connected component. CHREC_BEFORE is the
657 information that we already have collected up to this point.
658 TO_ADD is the evolution of "c".
660 When CHREC_BEFORE has an evolution part in LOOP_NB, add to this
661 evolution the expression TO_ADD, otherwise construct an evolution
662 part for this loop. */
664 static tree
665 add_to_evolution_1 (unsigned loop_nb, tree chrec_before, tree to_add,
666 gimple at_stmt)
668 tree type, left, right;
669 struct loop *loop = get_loop (loop_nb), *chloop;
671 switch (TREE_CODE (chrec_before))
673 case POLYNOMIAL_CHREC:
674 chloop = get_chrec_loop (chrec_before);
675 if (chloop == loop
676 || flow_loop_nested_p (chloop, loop))
678 unsigned var;
680 type = chrec_type (chrec_before);
682 /* When there is no evolution part in this loop, build it. */
683 if (chloop != loop)
685 var = loop_nb;
686 left = chrec_before;
687 right = SCALAR_FLOAT_TYPE_P (type)
688 ? build_real (type, dconst0)
689 : build_int_cst (type, 0);
691 else
693 var = CHREC_VARIABLE (chrec_before);
694 left = CHREC_LEFT (chrec_before);
695 right = CHREC_RIGHT (chrec_before);
698 to_add = chrec_convert (type, to_add, at_stmt);
699 right = chrec_convert_rhs (type, right, at_stmt);
700 right = chrec_fold_plus (chrec_type (right), right, to_add);
701 return build_polynomial_chrec (var, left, right);
703 else
705 gcc_assert (flow_loop_nested_p (loop, chloop));
707 /* Search the evolution in LOOP_NB. */
708 left = add_to_evolution_1 (loop_nb, CHREC_LEFT (chrec_before),
709 to_add, at_stmt);
710 right = CHREC_RIGHT (chrec_before);
711 right = chrec_convert_rhs (chrec_type (left), right, at_stmt);
712 return build_polynomial_chrec (CHREC_VARIABLE (chrec_before),
713 left, right);
716 default:
717 /* These nodes do not depend on a loop. */
718 if (chrec_before == chrec_dont_know)
719 return chrec_dont_know;
721 left = chrec_before;
722 right = chrec_convert_rhs (chrec_type (left), to_add, at_stmt);
723 return build_polynomial_chrec (loop_nb, left, right);
727 /* Add TO_ADD to the evolution part of CHREC_BEFORE in the dimension
728 of LOOP_NB.
730 Description (provided for completeness, for those who read code in
731 a plane, and for my poor 62 bytes brain that would have forgotten
732 all this in the next two or three months):
734 The algorithm of translation of programs from the SSA representation
735 into the chrecs syntax is based on a pattern matching. After having
736 reconstructed the overall tree expression for a loop, there are only
737 two cases that can arise:
739 1. a = loop-phi (init, a + expr)
740 2. a = loop-phi (init, expr)
742 where EXPR is either a scalar constant with respect to the analyzed
743 loop (this is a degree 0 polynomial), or an expression containing
744 other loop-phi definitions (these are higher degree polynomials).
746 Examples:
749 | init = ...
750 | loop_1
751 | a = phi (init, a + 5)
752 | endloop
755 | inita = ...
756 | initb = ...
757 | loop_1
758 | a = phi (inita, 2 * b + 3)
759 | b = phi (initb, b + 1)
760 | endloop
762 For the first case, the semantics of the SSA representation is:
764 | a (x) = init + \sum_{j = 0}^{x - 1} expr (j)
766 that is, there is a loop index "x" that determines the scalar value
767 of the variable during the loop execution. During the first
768 iteration, the value is that of the initial condition INIT, while
769 during the subsequent iterations, it is the sum of the initial
770 condition with the sum of all the values of EXPR from the initial
771 iteration to the before last considered iteration.
773 For the second case, the semantics of the SSA program is:
775 | a (x) = init, if x = 0;
776 | expr (x - 1), otherwise.
778 The second case corresponds to the PEELED_CHREC, whose syntax is
779 close to the syntax of a loop-phi-node:
781 | phi (init, expr) vs. (init, expr)_x
783 The proof of the translation algorithm for the first case is a
784 proof by structural induction based on the degree of EXPR.
786 Degree 0:
787 When EXPR is a constant with respect to the analyzed loop, or in
788 other words when EXPR is a polynomial of degree 0, the evolution of
789 the variable A in the loop is an affine function with an initial
790 condition INIT, and a step EXPR. In order to show this, we start
791 from the semantics of the SSA representation:
793 f (x) = init + \sum_{j = 0}^{x - 1} expr (j)
795 and since "expr (j)" is a constant with respect to "j",
797 f (x) = init + x * expr
799 Finally, based on the semantics of the pure sum chrecs, by
800 identification we get the corresponding chrecs syntax:
802 f (x) = init * \binom{x}{0} + expr * \binom{x}{1}
803 f (x) -> {init, +, expr}_x
805 Higher degree:
806 Suppose that EXPR is a polynomial of degree N with respect to the
807 analyzed loop_x for which we have already determined that it is
808 written under the chrecs syntax:
810 | expr (x) -> {b_0, +, b_1, +, ..., +, b_{n-1}} (x)
812 We start from the semantics of the SSA program:
814 | f (x) = init + \sum_{j = 0}^{x - 1} expr (j)
816 | f (x) = init + \sum_{j = 0}^{x - 1}
817 | (b_0 * \binom{j}{0} + ... + b_{n-1} * \binom{j}{n-1})
819 | f (x) = init + \sum_{j = 0}^{x - 1}
820 | \sum_{k = 0}^{n - 1} (b_k * \binom{j}{k})
822 | f (x) = init + \sum_{k = 0}^{n - 1}
823 | (b_k * \sum_{j = 0}^{x - 1} \binom{j}{k})
825 | f (x) = init + \sum_{k = 0}^{n - 1}
826 | (b_k * \binom{x}{k + 1})
828 | f (x) = init + b_0 * \binom{x}{1} + ...
829 | + b_{n-1} * \binom{x}{n}
831 | f (x) = init * \binom{x}{0} + b_0 * \binom{x}{1} + ...
832 | + b_{n-1} * \binom{x}{n}
835 And finally from the definition of the chrecs syntax, we identify:
836 | f (x) -> {init, +, b_0, +, ..., +, b_{n-1}}_x
838 This shows the mechanism that stands behind the add_to_evolution
839 function. An important point is that the use of symbolic
840 parameters avoids the need of an analysis schedule.
842 Example:
844 | inita = ...
845 | initb = ...
846 | loop_1
847 | a = phi (inita, a + 2 + b)
848 | b = phi (initb, b + 1)
849 | endloop
851 When analyzing "a", the algorithm keeps "b" symbolically:
853 | a -> {inita, +, 2 + b}_1
855 Then, after instantiation, the analyzer ends on the evolution:
857 | a -> {inita, +, 2 + initb, +, 1}_1
861 static tree
862 add_to_evolution (unsigned loop_nb, tree chrec_before, enum tree_code code,
863 tree to_add, gimple at_stmt)
865 tree type = chrec_type (to_add);
866 tree res = NULL_TREE;
868 if (to_add == NULL_TREE)
869 return chrec_before;
871 /* TO_ADD is either a scalar, or a parameter. TO_ADD is not
872 instantiated at this point. */
873 if (TREE_CODE (to_add) == POLYNOMIAL_CHREC)
874 /* This should not happen. */
875 return chrec_dont_know;
877 if (dump_file && (dump_flags & TDF_DETAILS))
879 fprintf (dump_file, "(add_to_evolution \n");
880 fprintf (dump_file, " (loop_nb = %d)\n", loop_nb);
881 fprintf (dump_file, " (chrec_before = ");
882 print_generic_expr (dump_file, chrec_before, 0);
883 fprintf (dump_file, ")\n (to_add = ");
884 print_generic_expr (dump_file, to_add, 0);
885 fprintf (dump_file, ")\n");
888 if (code == MINUS_EXPR)
889 to_add = chrec_fold_multiply (type, to_add, SCALAR_FLOAT_TYPE_P (type)
890 ? build_real (type, dconstm1)
891 : build_int_cst_type (type, -1));
893 res = add_to_evolution_1 (loop_nb, chrec_before, to_add, at_stmt);
895 if (dump_file && (dump_flags & TDF_DETAILS))
897 fprintf (dump_file, " (res = ");
898 print_generic_expr (dump_file, res, 0);
899 fprintf (dump_file, "))\n");
902 return res;
905 /* Helper function. */
907 static inline tree
908 set_nb_iterations_in_loop (struct loop *loop,
909 tree res)
911 if (dump_file && (dump_flags & TDF_DETAILS))
913 fprintf (dump_file, " (set_nb_iterations_in_loop = ");
914 print_generic_expr (dump_file, res, 0);
915 fprintf (dump_file, "))\n");
918 loop->nb_iterations = res;
919 return res;
924 /* This section selects the loops that will be good candidates for the
925 scalar evolution analysis. For the moment, greedily select all the
926 loop nests we could analyze. */
928 /* For a loop with a single exit edge, return the COND_EXPR that
929 guards the exit edge. If the expression is too difficult to
930 analyze, then give up. */
932 gimple
933 get_loop_exit_condition (const struct loop *loop)
935 gimple res = NULL;
936 edge exit_edge = single_exit (loop);
938 if (dump_file && (dump_flags & TDF_DETAILS))
939 fprintf (dump_file, "(get_loop_exit_condition \n ");
941 if (exit_edge)
943 gimple stmt;
945 stmt = last_stmt (exit_edge->src);
946 if (gimple_code (stmt) == GIMPLE_COND)
947 res = stmt;
950 if (dump_file && (dump_flags & TDF_DETAILS))
952 print_gimple_stmt (dump_file, res, 0, 0);
953 fprintf (dump_file, ")\n");
956 return res;
959 /* Recursively determine and enqueue the exit conditions for a loop. */
961 static void
962 get_exit_conditions_rec (struct loop *loop,
963 VEC(gimple,heap) **exit_conditions)
965 if (!loop)
966 return;
968 /* Recurse on the inner loops, then on the next (sibling) loops. */
969 get_exit_conditions_rec (loop->inner, exit_conditions);
970 get_exit_conditions_rec (loop->next, exit_conditions);
972 if (single_exit (loop))
974 gimple loop_condition = get_loop_exit_condition (loop);
976 if (loop_condition)
977 VEC_safe_push (gimple, heap, *exit_conditions, loop_condition);
981 /* Select the candidate loop nests for the analysis. This function
982 initializes the EXIT_CONDITIONS array. */
984 static void
985 select_loops_exit_conditions (VEC(gimple,heap) **exit_conditions)
987 struct loop *function_body = current_loops->tree_root;
989 get_exit_conditions_rec (function_body->inner, exit_conditions);
993 /* Depth first search algorithm. */
995 typedef enum t_bool {
996 t_false,
997 t_true,
998 t_dont_know
999 } t_bool;
1002 static t_bool follow_ssa_edge (struct loop *loop, gimple, gimple, tree *, int);
1004 /* Follow the ssa edge into the binary expression RHS0 CODE RHS1.
1005 Return true if the strongly connected component has been found. */
1007 static t_bool
1008 follow_ssa_edge_binary (struct loop *loop, gimple at_stmt,
1009 tree type, tree rhs0, enum tree_code code, tree rhs1,
1010 gimple halting_phi, tree *evolution_of_loop, int limit)
1012 t_bool res = t_false;
1013 tree evol;
1015 switch (code)
1017 case POINTER_PLUS_EXPR:
1018 case PLUS_EXPR:
1019 if (TREE_CODE (rhs0) == SSA_NAME)
1021 if (TREE_CODE (rhs1) == SSA_NAME)
1023 /* Match an assignment under the form:
1024 "a = b + c". */
1026 /* We want only assignments of form "name + name" contribute to
1027 LIMIT, as the other cases do not necessarily contribute to
1028 the complexity of the expression. */
1029 limit++;
1031 evol = *evolution_of_loop;
1032 res = follow_ssa_edge
1033 (loop, SSA_NAME_DEF_STMT (rhs0), halting_phi, &evol, limit);
1035 if (res == t_true)
1036 *evolution_of_loop = add_to_evolution
1037 (loop->num,
1038 chrec_convert (type, evol, at_stmt),
1039 code, rhs1, at_stmt);
1041 else if (res == t_false)
1043 res = follow_ssa_edge
1044 (loop, SSA_NAME_DEF_STMT (rhs1), halting_phi,
1045 evolution_of_loop, limit);
1047 if (res == t_true)
1048 *evolution_of_loop = add_to_evolution
1049 (loop->num,
1050 chrec_convert (type, *evolution_of_loop, at_stmt),
1051 code, rhs0, at_stmt);
1053 else if (res == t_dont_know)
1054 *evolution_of_loop = chrec_dont_know;
1057 else if (res == t_dont_know)
1058 *evolution_of_loop = chrec_dont_know;
1061 else
1063 /* Match an assignment under the form:
1064 "a = b + ...". */
1065 res = follow_ssa_edge
1066 (loop, SSA_NAME_DEF_STMT (rhs0), halting_phi,
1067 evolution_of_loop, limit);
1068 if (res == t_true)
1069 *evolution_of_loop = add_to_evolution
1070 (loop->num, chrec_convert (type, *evolution_of_loop,
1071 at_stmt),
1072 code, rhs1, at_stmt);
1074 else if (res == t_dont_know)
1075 *evolution_of_loop = chrec_dont_know;
1079 else if (TREE_CODE (rhs1) == SSA_NAME)
1081 /* Match an assignment under the form:
1082 "a = ... + c". */
1083 res = follow_ssa_edge
1084 (loop, SSA_NAME_DEF_STMT (rhs1), halting_phi,
1085 evolution_of_loop, limit);
1086 if (res == t_true)
1087 *evolution_of_loop = add_to_evolution
1088 (loop->num, chrec_convert (type, *evolution_of_loop,
1089 at_stmt),
1090 code, rhs0, at_stmt);
1092 else if (res == t_dont_know)
1093 *evolution_of_loop = chrec_dont_know;
1096 else
1097 /* Otherwise, match an assignment under the form:
1098 "a = ... + ...". */
1099 /* And there is nothing to do. */
1100 res = t_false;
1101 break;
1103 case MINUS_EXPR:
1104 /* This case is under the form "opnd0 = rhs0 - rhs1". */
1105 if (TREE_CODE (rhs0) == SSA_NAME)
1107 /* Match an assignment under the form:
1108 "a = b - ...". */
1110 /* We want only assignments of form "name - name" contribute to
1111 LIMIT, as the other cases do not necessarily contribute to
1112 the complexity of the expression. */
1113 if (TREE_CODE (rhs1) == SSA_NAME)
1114 limit++;
1116 res = follow_ssa_edge (loop, SSA_NAME_DEF_STMT (rhs0), halting_phi,
1117 evolution_of_loop, limit);
1118 if (res == t_true)
1119 *evolution_of_loop = add_to_evolution
1120 (loop->num, chrec_convert (type, *evolution_of_loop, at_stmt),
1121 MINUS_EXPR, rhs1, at_stmt);
1123 else if (res == t_dont_know)
1124 *evolution_of_loop = chrec_dont_know;
1126 else
1127 /* Otherwise, match an assignment under the form:
1128 "a = ... - ...". */
1129 /* And there is nothing to do. */
1130 res = t_false;
1131 break;
1133 default:
1134 res = t_false;
1137 return res;
1140 /* Follow the ssa edge into the expression EXPR.
1141 Return true if the strongly connected component has been found. */
1143 static t_bool
1144 follow_ssa_edge_expr (struct loop *loop, gimple at_stmt, tree expr,
1145 gimple halting_phi, tree *evolution_of_loop, int limit)
1147 enum tree_code code = TREE_CODE (expr);
1148 tree type = TREE_TYPE (expr), rhs0, rhs1;
1149 t_bool res;
1151 /* The EXPR is one of the following cases:
1152 - an SSA_NAME,
1153 - an INTEGER_CST,
1154 - a PLUS_EXPR,
1155 - a POINTER_PLUS_EXPR,
1156 - a MINUS_EXPR,
1157 - an ASSERT_EXPR,
1158 - other cases are not yet handled. */
1160 switch (code)
1162 CASE_CONVERT:
1163 /* This assignment is under the form "a_1 = (cast) rhs. */
1164 res = follow_ssa_edge_expr (loop, at_stmt, TREE_OPERAND (expr, 0),
1165 halting_phi, evolution_of_loop, limit);
1166 *evolution_of_loop = chrec_convert (type, *evolution_of_loop, at_stmt);
1167 break;
1169 case INTEGER_CST:
1170 /* This assignment is under the form "a_1 = 7". */
1171 res = t_false;
1172 break;
1174 case SSA_NAME:
1175 /* This assignment is under the form: "a_1 = b_2". */
1176 res = follow_ssa_edge
1177 (loop, SSA_NAME_DEF_STMT (expr), halting_phi, evolution_of_loop, limit);
1178 break;
1180 case POINTER_PLUS_EXPR:
1181 case PLUS_EXPR:
1182 case MINUS_EXPR:
1183 /* This case is under the form "rhs0 +- rhs1". */
1184 rhs0 = TREE_OPERAND (expr, 0);
1185 rhs1 = TREE_OPERAND (expr, 1);
1186 type = TREE_TYPE (rhs0);
1187 STRIP_USELESS_TYPE_CONVERSION (rhs0);
1188 STRIP_USELESS_TYPE_CONVERSION (rhs1);
1189 res = follow_ssa_edge_binary (loop, at_stmt, type, rhs0, code, rhs1,
1190 halting_phi, evolution_of_loop, limit);
1191 break;
1193 case ASSERT_EXPR:
1194 /* This assignment is of the form: "a_1 = ASSERT_EXPR <a_2, ...>"
1195 It must be handled as a copy assignment of the form a_1 = a_2. */
1196 rhs0 = ASSERT_EXPR_VAR (expr);
1197 if (TREE_CODE (rhs0) == SSA_NAME)
1198 res = follow_ssa_edge (loop, SSA_NAME_DEF_STMT (rhs0),
1199 halting_phi, evolution_of_loop, limit);
1200 else
1201 res = t_false;
1202 break;
1204 default:
1205 res = t_false;
1206 break;
1209 return res;
1212 /* Follow the ssa edge into the right hand side of an assignment STMT.
1213 Return true if the strongly connected component has been found. */
1215 static t_bool
1216 follow_ssa_edge_in_rhs (struct loop *loop, gimple stmt,
1217 gimple halting_phi, tree *evolution_of_loop, int limit)
1219 enum tree_code code = gimple_assign_rhs_code (stmt);
1220 tree type = gimple_expr_type (stmt), rhs1, rhs2;
1221 t_bool res;
1223 switch (code)
1225 CASE_CONVERT:
1226 /* This assignment is under the form "a_1 = (cast) rhs. */
1227 res = follow_ssa_edge_expr (loop, stmt, gimple_assign_rhs1 (stmt),
1228 halting_phi, evolution_of_loop, limit);
1229 *evolution_of_loop = chrec_convert (type, *evolution_of_loop, stmt);
1230 break;
1232 case POINTER_PLUS_EXPR:
1233 case PLUS_EXPR:
1234 case MINUS_EXPR:
1235 rhs1 = gimple_assign_rhs1 (stmt);
1236 rhs2 = gimple_assign_rhs2 (stmt);
1237 type = TREE_TYPE (rhs1);
1238 res = follow_ssa_edge_binary (loop, stmt, type, rhs1, code, rhs2,
1239 halting_phi, evolution_of_loop, limit);
1240 break;
1242 default:
1243 if (get_gimple_rhs_class (code) == GIMPLE_SINGLE_RHS)
1244 res = follow_ssa_edge_expr (loop, stmt, gimple_assign_rhs1 (stmt),
1245 halting_phi, evolution_of_loop, limit);
1246 else
1247 res = t_false;
1248 break;
1251 return res;
1254 /* Checks whether the I-th argument of a PHI comes from a backedge. */
1256 static bool
1257 backedge_phi_arg_p (gimple phi, int i)
1259 const_edge e = gimple_phi_arg_edge (phi, i);
1261 /* We would in fact like to test EDGE_DFS_BACK here, but we do not care
1262 about updating it anywhere, and this should work as well most of the
1263 time. */
1264 if (e->flags & EDGE_IRREDUCIBLE_LOOP)
1265 return true;
1267 return false;
1270 /* Helper function for one branch of the condition-phi-node. Return
1271 true if the strongly connected component has been found following
1272 this path. */
1274 static inline t_bool
1275 follow_ssa_edge_in_condition_phi_branch (int i,
1276 struct loop *loop,
1277 gimple condition_phi,
1278 gimple halting_phi,
1279 tree *evolution_of_branch,
1280 tree init_cond, int limit)
1282 tree branch = PHI_ARG_DEF (condition_phi, i);
1283 *evolution_of_branch = chrec_dont_know;
1285 /* Do not follow back edges (they must belong to an irreducible loop, which
1286 we really do not want to worry about). */
1287 if (backedge_phi_arg_p (condition_phi, i))
1288 return t_false;
1290 if (TREE_CODE (branch) == SSA_NAME)
1292 *evolution_of_branch = init_cond;
1293 return follow_ssa_edge (loop, SSA_NAME_DEF_STMT (branch), halting_phi,
1294 evolution_of_branch, limit);
1297 /* This case occurs when one of the condition branches sets
1298 the variable to a constant: i.e. a phi-node like
1299 "a_2 = PHI <a_7(5), 2(6)>;".
1301 FIXME: This case have to be refined correctly:
1302 in some cases it is possible to say something better than
1303 chrec_dont_know, for example using a wrap-around notation. */
1304 return t_false;
1307 /* This function merges the branches of a condition-phi-node in a
1308 loop. */
1310 static t_bool
1311 follow_ssa_edge_in_condition_phi (struct loop *loop,
1312 gimple condition_phi,
1313 gimple halting_phi,
1314 tree *evolution_of_loop, int limit)
1316 int i, n;
1317 tree init = *evolution_of_loop;
1318 tree evolution_of_branch;
1319 t_bool res = follow_ssa_edge_in_condition_phi_branch (0, loop, condition_phi,
1320 halting_phi,
1321 &evolution_of_branch,
1322 init, limit);
1323 if (res == t_false || res == t_dont_know)
1324 return res;
1326 *evolution_of_loop = evolution_of_branch;
1328 n = gimple_phi_num_args (condition_phi);
1329 for (i = 1; i < n; i++)
1331 /* Quickly give up when the evolution of one of the branches is
1332 not known. */
1333 if (*evolution_of_loop == chrec_dont_know)
1334 return t_true;
1336 /* Increase the limit by the PHI argument number to avoid exponential
1337 time and memory complexity. */
1338 res = follow_ssa_edge_in_condition_phi_branch (i, loop, condition_phi,
1339 halting_phi,
1340 &evolution_of_branch,
1341 init, limit + i);
1342 if (res == t_false || res == t_dont_know)
1343 return res;
1345 *evolution_of_loop = chrec_merge (*evolution_of_loop,
1346 evolution_of_branch);
1349 return t_true;
1352 /* Follow an SSA edge in an inner loop. It computes the overall
1353 effect of the loop, and following the symbolic initial conditions,
1354 it follows the edges in the parent loop. The inner loop is
1355 considered as a single statement. */
1357 static t_bool
1358 follow_ssa_edge_inner_loop_phi (struct loop *outer_loop,
1359 gimple loop_phi_node,
1360 gimple halting_phi,
1361 tree *evolution_of_loop, int limit)
1363 struct loop *loop = loop_containing_stmt (loop_phi_node);
1364 tree ev = analyze_scalar_evolution (loop, PHI_RESULT (loop_phi_node));
1366 /* Sometimes, the inner loop is too difficult to analyze, and the
1367 result of the analysis is a symbolic parameter. */
1368 if (ev == PHI_RESULT (loop_phi_node))
1370 t_bool res = t_false;
1371 int i, n = gimple_phi_num_args (loop_phi_node);
1373 for (i = 0; i < n; i++)
1375 tree arg = PHI_ARG_DEF (loop_phi_node, i);
1376 basic_block bb;
1378 /* Follow the edges that exit the inner loop. */
1379 bb = gimple_phi_arg_edge (loop_phi_node, i)->src;
1380 if (!flow_bb_inside_loop_p (loop, bb))
1381 res = follow_ssa_edge_expr (outer_loop, loop_phi_node,
1382 arg, halting_phi,
1383 evolution_of_loop, limit);
1384 if (res == t_true)
1385 break;
1388 /* If the path crosses this loop-phi, give up. */
1389 if (res == t_true)
1390 *evolution_of_loop = chrec_dont_know;
1392 return res;
1395 /* Otherwise, compute the overall effect of the inner loop. */
1396 ev = compute_overall_effect_of_inner_loop (loop, ev);
1397 return follow_ssa_edge_expr (outer_loop, loop_phi_node, ev, halting_phi,
1398 evolution_of_loop, limit);
1401 /* Follow an SSA edge from a loop-phi-node to itself, constructing a
1402 path that is analyzed on the return walk. */
1404 static t_bool
1405 follow_ssa_edge (struct loop *loop, gimple def, gimple halting_phi,
1406 tree *evolution_of_loop, int limit)
1408 struct loop *def_loop;
1410 if (gimple_nop_p (def))
1411 return t_false;
1413 /* Give up if the path is longer than the MAX that we allow. */
1414 if (limit > PARAM_VALUE (PARAM_SCEV_MAX_EXPR_SIZE))
1415 return t_dont_know;
1417 def_loop = loop_containing_stmt (def);
1419 switch (gimple_code (def))
1421 case GIMPLE_PHI:
1422 if (!loop_phi_node_p (def))
1423 /* DEF is a condition-phi-node. Follow the branches, and
1424 record their evolutions. Finally, merge the collected
1425 information and set the approximation to the main
1426 variable. */
1427 return follow_ssa_edge_in_condition_phi
1428 (loop, def, halting_phi, evolution_of_loop, limit);
1430 /* When the analyzed phi is the halting_phi, the
1431 depth-first search is over: we have found a path from
1432 the halting_phi to itself in the loop. */
1433 if (def == halting_phi)
1434 return t_true;
1436 /* Otherwise, the evolution of the HALTING_PHI depends
1437 on the evolution of another loop-phi-node, i.e. the
1438 evolution function is a higher degree polynomial. */
1439 if (def_loop == loop)
1440 return t_false;
1442 /* Inner loop. */
1443 if (flow_loop_nested_p (loop, def_loop))
1444 return follow_ssa_edge_inner_loop_phi
1445 (loop, def, halting_phi, evolution_of_loop, limit + 1);
1447 /* Outer loop. */
1448 return t_false;
1450 case GIMPLE_ASSIGN:
1451 return follow_ssa_edge_in_rhs (loop, def, halting_phi,
1452 evolution_of_loop, limit);
1454 default:
1455 /* At this level of abstraction, the program is just a set
1456 of GIMPLE_ASSIGNs and PHI_NODEs. In principle there is no
1457 other node to be handled. */
1458 return t_false;
1464 /* Given a LOOP_PHI_NODE, this function determines the evolution
1465 function from LOOP_PHI_NODE to LOOP_PHI_NODE in the loop. */
1467 static tree
1468 analyze_evolution_in_loop (gimple loop_phi_node,
1469 tree init_cond)
1471 int i, n = gimple_phi_num_args (loop_phi_node);
1472 tree evolution_function = chrec_not_analyzed_yet;
1473 struct loop *loop = loop_containing_stmt (loop_phi_node);
1474 basic_block bb;
1476 if (dump_file && (dump_flags & TDF_DETAILS))
1478 fprintf (dump_file, "(analyze_evolution_in_loop \n");
1479 fprintf (dump_file, " (loop_phi_node = ");
1480 print_gimple_stmt (dump_file, loop_phi_node, 0, 0);
1481 fprintf (dump_file, ")\n");
1484 for (i = 0; i < n; i++)
1486 tree arg = PHI_ARG_DEF (loop_phi_node, i);
1487 gimple ssa_chain;
1488 tree ev_fn;
1489 t_bool res;
1491 /* Select the edges that enter the loop body. */
1492 bb = gimple_phi_arg_edge (loop_phi_node, i)->src;
1493 if (!flow_bb_inside_loop_p (loop, bb))
1494 continue;
1496 if (TREE_CODE (arg) == SSA_NAME)
1498 ssa_chain = SSA_NAME_DEF_STMT (arg);
1500 /* Pass in the initial condition to the follow edge function. */
1501 ev_fn = init_cond;
1502 res = follow_ssa_edge (loop, ssa_chain, loop_phi_node, &ev_fn, 0);
1504 else
1505 res = t_false;
1507 /* When it is impossible to go back on the same
1508 loop_phi_node by following the ssa edges, the
1509 evolution is represented by a peeled chrec, i.e. the
1510 first iteration, EV_FN has the value INIT_COND, then
1511 all the other iterations it has the value of ARG.
1512 For the moment, PEELED_CHREC nodes are not built. */
1513 if (res != t_true)
1514 ev_fn = chrec_dont_know;
1516 /* When there are multiple back edges of the loop (which in fact never
1517 happens currently, but nevertheless), merge their evolutions. */
1518 evolution_function = chrec_merge (evolution_function, ev_fn);
1521 if (dump_file && (dump_flags & TDF_DETAILS))
1523 fprintf (dump_file, " (evolution_function = ");
1524 print_generic_expr (dump_file, evolution_function, 0);
1525 fprintf (dump_file, "))\n");
1528 return evolution_function;
1531 /* Given a loop-phi-node, return the initial conditions of the
1532 variable on entry of the loop. When the CCP has propagated
1533 constants into the loop-phi-node, the initial condition is
1534 instantiated, otherwise the initial condition is kept symbolic.
1535 This analyzer does not analyze the evolution outside the current
1536 loop, and leaves this task to the on-demand tree reconstructor. */
1538 static tree
1539 analyze_initial_condition (gimple loop_phi_node)
1541 int i, n;
1542 tree init_cond = chrec_not_analyzed_yet;
1543 struct loop *loop = loop_containing_stmt (loop_phi_node);
1545 if (dump_file && (dump_flags & TDF_DETAILS))
1547 fprintf (dump_file, "(analyze_initial_condition \n");
1548 fprintf (dump_file, " (loop_phi_node = \n");
1549 print_gimple_stmt (dump_file, loop_phi_node, 0, 0);
1550 fprintf (dump_file, ")\n");
1553 n = gimple_phi_num_args (loop_phi_node);
1554 for (i = 0; i < n; i++)
1556 tree branch = PHI_ARG_DEF (loop_phi_node, i);
1557 basic_block bb = gimple_phi_arg_edge (loop_phi_node, i)->src;
1559 /* When the branch is oriented to the loop's body, it does
1560 not contribute to the initial condition. */
1561 if (flow_bb_inside_loop_p (loop, bb))
1562 continue;
1564 if (init_cond == chrec_not_analyzed_yet)
1566 init_cond = branch;
1567 continue;
1570 if (TREE_CODE (branch) == SSA_NAME)
1572 init_cond = chrec_dont_know;
1573 break;
1576 init_cond = chrec_merge (init_cond, branch);
1579 /* Ooops -- a loop without an entry??? */
1580 if (init_cond == chrec_not_analyzed_yet)
1581 init_cond = chrec_dont_know;
1583 /* During early loop unrolling we do not have fully constant propagated IL.
1584 Handle degenerate PHIs here to not miss important unrollings. */
1585 if (TREE_CODE (init_cond) == SSA_NAME)
1587 gimple def = SSA_NAME_DEF_STMT (init_cond);
1588 tree res;
1589 if (gimple_code (def) == GIMPLE_PHI
1590 && (res = degenerate_phi_result (def)) != NULL_TREE
1591 /* Only allow invariants here, otherwise we may break
1592 loop-closed SSA form. */
1593 && is_gimple_min_invariant (res))
1594 init_cond = res;
1597 if (dump_file && (dump_flags & TDF_DETAILS))
1599 fprintf (dump_file, " (init_cond = ");
1600 print_generic_expr (dump_file, init_cond, 0);
1601 fprintf (dump_file, "))\n");
1604 return init_cond;
1607 /* Analyze the scalar evolution for LOOP_PHI_NODE. */
1609 static tree
1610 interpret_loop_phi (struct loop *loop, gimple loop_phi_node)
1612 tree res;
1613 struct loop *phi_loop = loop_containing_stmt (loop_phi_node);
1614 tree init_cond;
1616 if (phi_loop != loop)
1618 struct loop *subloop;
1619 tree evolution_fn = analyze_scalar_evolution
1620 (phi_loop, PHI_RESULT (loop_phi_node));
1622 /* Dive one level deeper. */
1623 subloop = superloop_at_depth (phi_loop, loop_depth (loop) + 1);
1625 /* Interpret the subloop. */
1626 res = compute_overall_effect_of_inner_loop (subloop, evolution_fn);
1627 return res;
1630 /* Otherwise really interpret the loop phi. */
1631 init_cond = analyze_initial_condition (loop_phi_node);
1632 res = analyze_evolution_in_loop (loop_phi_node, init_cond);
1634 return res;
1637 /* This function merges the branches of a condition-phi-node,
1638 contained in the outermost loop, and whose arguments are already
1639 analyzed. */
1641 static tree
1642 interpret_condition_phi (struct loop *loop, gimple condition_phi)
1644 int i, n = gimple_phi_num_args (condition_phi);
1645 tree res = chrec_not_analyzed_yet;
1647 for (i = 0; i < n; i++)
1649 tree branch_chrec;
1651 if (backedge_phi_arg_p (condition_phi, i))
1653 res = chrec_dont_know;
1654 break;
1657 branch_chrec = analyze_scalar_evolution
1658 (loop, PHI_ARG_DEF (condition_phi, i));
1660 res = chrec_merge (res, branch_chrec);
1663 return res;
1666 /* Interpret the operation RHS1 OP RHS2. If we didn't
1667 analyze this node before, follow the definitions until ending
1668 either on an analyzed GIMPLE_ASSIGN, or on a loop-phi-node. On the
1669 return path, this function propagates evolutions (ala constant copy
1670 propagation). OPND1 is not a GIMPLE expression because we could
1671 analyze the effect of an inner loop: see interpret_loop_phi. */
1673 static tree
1674 interpret_rhs_expr (struct loop *loop, gimple at_stmt,
1675 tree type, tree rhs1, enum tree_code code, tree rhs2)
1677 tree res, chrec1, chrec2;
1679 if (get_gimple_rhs_class (code) == GIMPLE_SINGLE_RHS)
1681 if (is_gimple_min_invariant (rhs1))
1682 return chrec_convert (type, rhs1, at_stmt);
1684 if (code == SSA_NAME)
1685 return chrec_convert (type, analyze_scalar_evolution (loop, rhs1),
1686 at_stmt);
1688 if (code == ASSERT_EXPR)
1690 rhs1 = ASSERT_EXPR_VAR (rhs1);
1691 return chrec_convert (type, analyze_scalar_evolution (loop, rhs1),
1692 at_stmt);
1695 return chrec_dont_know;
1698 switch (code)
1700 case POINTER_PLUS_EXPR:
1701 chrec1 = analyze_scalar_evolution (loop, rhs1);
1702 chrec2 = analyze_scalar_evolution (loop, rhs2);
1703 chrec1 = chrec_convert (type, chrec1, at_stmt);
1704 chrec2 = chrec_convert (sizetype, chrec2, at_stmt);
1705 res = chrec_fold_plus (type, chrec1, chrec2);
1706 break;
1708 case PLUS_EXPR:
1709 chrec1 = analyze_scalar_evolution (loop, rhs1);
1710 chrec2 = analyze_scalar_evolution (loop, rhs2);
1711 chrec1 = chrec_convert (type, chrec1, at_stmt);
1712 chrec2 = chrec_convert (type, chrec2, at_stmt);
1713 res = chrec_fold_plus (type, chrec1, chrec2);
1714 break;
1716 case MINUS_EXPR:
1717 chrec1 = analyze_scalar_evolution (loop, rhs1);
1718 chrec2 = analyze_scalar_evolution (loop, rhs2);
1719 chrec1 = chrec_convert (type, chrec1, at_stmt);
1720 chrec2 = chrec_convert (type, chrec2, at_stmt);
1721 res = chrec_fold_minus (type, chrec1, chrec2);
1722 break;
1724 case NEGATE_EXPR:
1725 chrec1 = analyze_scalar_evolution (loop, rhs1);
1726 chrec1 = chrec_convert (type, chrec1, at_stmt);
1727 /* TYPE may be integer, real or complex, so use fold_convert. */
1728 res = chrec_fold_multiply (type, chrec1,
1729 fold_convert (type, integer_minus_one_node));
1730 break;
1732 case BIT_NOT_EXPR:
1733 /* Handle ~X as -1 - X. */
1734 chrec1 = analyze_scalar_evolution (loop, rhs1);
1735 chrec1 = chrec_convert (type, chrec1, at_stmt);
1736 res = chrec_fold_minus (type,
1737 fold_convert (type, integer_minus_one_node),
1738 chrec1);
1739 break;
1741 case MULT_EXPR:
1742 chrec1 = analyze_scalar_evolution (loop, rhs1);
1743 chrec2 = analyze_scalar_evolution (loop, rhs2);
1744 chrec1 = chrec_convert (type, chrec1, at_stmt);
1745 chrec2 = chrec_convert (type, chrec2, at_stmt);
1746 res = chrec_fold_multiply (type, chrec1, chrec2);
1747 break;
1749 CASE_CONVERT:
1750 chrec1 = analyze_scalar_evolution (loop, rhs1);
1751 res = chrec_convert (type, chrec1, at_stmt);
1752 break;
1754 default:
1755 res = chrec_dont_know;
1756 break;
1759 return res;
1762 /* Interpret the expression EXPR. */
1764 static tree
1765 interpret_expr (struct loop *loop, gimple at_stmt, tree expr)
1767 enum tree_code code;
1768 tree type = TREE_TYPE (expr), op0, op1;
1770 if (automatically_generated_chrec_p (expr))
1771 return expr;
1773 if (TREE_CODE (expr) == POLYNOMIAL_CHREC)
1774 return chrec_dont_know;
1776 extract_ops_from_tree (expr, &code, &op0, &op1);
1778 return interpret_rhs_expr (loop, at_stmt, type,
1779 op0, code, op1);
1782 /* Interpret the rhs of the assignment STMT. */
1784 static tree
1785 interpret_gimple_assign (struct loop *loop, gimple stmt)
1787 tree type = TREE_TYPE (gimple_assign_lhs (stmt));
1788 enum tree_code code = gimple_assign_rhs_code (stmt);
1790 return interpret_rhs_expr (loop, stmt, type,
1791 gimple_assign_rhs1 (stmt), code,
1792 gimple_assign_rhs2 (stmt));
1797 /* This section contains all the entry points:
1798 - number_of_iterations_in_loop,
1799 - analyze_scalar_evolution,
1800 - instantiate_parameters.
1803 /* Compute and return the evolution function in WRTO_LOOP, the nearest
1804 common ancestor of DEF_LOOP and USE_LOOP. */
1806 static tree
1807 compute_scalar_evolution_in_loop (struct loop *wrto_loop,
1808 struct loop *def_loop,
1809 tree ev)
1811 tree res;
1812 if (def_loop == wrto_loop)
1813 return ev;
1815 def_loop = superloop_at_depth (def_loop, loop_depth (wrto_loop) + 1);
1816 res = compute_overall_effect_of_inner_loop (def_loop, ev);
1818 return analyze_scalar_evolution_1 (wrto_loop, res, chrec_not_analyzed_yet);
1821 /* Helper recursive function. */
1823 static tree
1824 analyze_scalar_evolution_1 (struct loop *loop, tree var, tree res)
1826 tree type = TREE_TYPE (var);
1827 gimple def;
1828 basic_block bb;
1829 struct loop *def_loop;
1831 if (loop == NULL || TREE_CODE (type) == VECTOR_TYPE)
1832 return chrec_dont_know;
1834 if (TREE_CODE (var) != SSA_NAME)
1835 return interpret_expr (loop, NULL, var);
1837 def = SSA_NAME_DEF_STMT (var);
1838 bb = gimple_bb (def);
1839 def_loop = bb ? bb->loop_father : NULL;
1841 if (bb == NULL
1842 || !flow_bb_inside_loop_p (loop, bb))
1844 /* Keep the symbolic form. */
1845 res = var;
1846 goto set_and_end;
1849 if (res != chrec_not_analyzed_yet)
1851 if (loop != bb->loop_father)
1852 res = compute_scalar_evolution_in_loop
1853 (find_common_loop (loop, bb->loop_father), bb->loop_father, res);
1855 goto set_and_end;
1858 if (loop != def_loop)
1860 res = analyze_scalar_evolution_1 (def_loop, var, chrec_not_analyzed_yet);
1861 res = compute_scalar_evolution_in_loop (loop, def_loop, res);
1863 goto set_and_end;
1866 switch (gimple_code (def))
1868 case GIMPLE_ASSIGN:
1869 res = interpret_gimple_assign (loop, def);
1870 break;
1872 case GIMPLE_PHI:
1873 if (loop_phi_node_p (def))
1874 res = interpret_loop_phi (loop, def);
1875 else
1876 res = interpret_condition_phi (loop, def);
1877 break;
1879 default:
1880 res = chrec_dont_know;
1881 break;
1884 set_and_end:
1886 /* Keep the symbolic form. */
1887 if (res == chrec_dont_know)
1888 res = var;
1890 if (loop == def_loop)
1891 set_scalar_evolution (block_before_loop (loop), var, res);
1893 return res;
1896 /* Analyzes and returns the scalar evolution of the ssa_name VAR in
1897 LOOP. LOOP is the loop in which the variable is used.
1899 Example of use: having a pointer VAR to a SSA_NAME node, STMT a
1900 pointer to the statement that uses this variable, in order to
1901 determine the evolution function of the variable, use the following
1902 calls:
1904 loop_p loop = loop_containing_stmt (stmt);
1905 tree chrec_with_symbols = analyze_scalar_evolution (loop, var);
1906 tree chrec_instantiated = instantiate_parameters (loop, chrec_with_symbols);
1909 tree
1910 analyze_scalar_evolution (struct loop *loop, tree var)
1912 tree res;
1914 if (dump_file && (dump_flags & TDF_DETAILS))
1916 fprintf (dump_file, "(analyze_scalar_evolution \n");
1917 fprintf (dump_file, " (loop_nb = %d)\n", loop->num);
1918 fprintf (dump_file, " (scalar = ");
1919 print_generic_expr (dump_file, var, 0);
1920 fprintf (dump_file, ")\n");
1923 res = get_scalar_evolution (block_before_loop (loop), var);
1924 res = analyze_scalar_evolution_1 (loop, var, res);
1926 if (dump_file && (dump_flags & TDF_DETAILS))
1927 fprintf (dump_file, ")\n");
1929 return res;
1932 /* Analyze scalar evolution of use of VERSION in USE_LOOP with respect to
1933 WRTO_LOOP (which should be a superloop of USE_LOOP)
1935 FOLDED_CASTS is set to true if resolve_mixers used
1936 chrec_convert_aggressive (TODO -- not really, we are way too conservative
1937 at the moment in order to keep things simple).
1939 To illustrate the meaning of USE_LOOP and WRTO_LOOP, consider the following
1940 example:
1942 for (i = 0; i < 100; i++) -- loop 1
1944 for (j = 0; j < 100; j++) -- loop 2
1946 k1 = i;
1947 k2 = j;
1949 use2 (k1, k2);
1951 for (t = 0; t < 100; t++) -- loop 3
1952 use3 (k1, k2);
1955 use1 (k1, k2);
1958 Both k1 and k2 are invariants in loop3, thus
1959 analyze_scalar_evolution_in_loop (loop3, loop3, k1) = k1
1960 analyze_scalar_evolution_in_loop (loop3, loop3, k2) = k2
1962 As they are invariant, it does not matter whether we consider their
1963 usage in loop 3 or loop 2, hence
1964 analyze_scalar_evolution_in_loop (loop2, loop3, k1) =
1965 analyze_scalar_evolution_in_loop (loop2, loop2, k1) = i
1966 analyze_scalar_evolution_in_loop (loop2, loop3, k2) =
1967 analyze_scalar_evolution_in_loop (loop2, loop2, k2) = [0,+,1]_2
1969 Similarly for their evolutions with respect to loop 1. The values of K2
1970 in the use in loop 2 vary independently on loop 1, thus we cannot express
1971 the evolution with respect to loop 1:
1972 analyze_scalar_evolution_in_loop (loop1, loop3, k1) =
1973 analyze_scalar_evolution_in_loop (loop1, loop2, k1) = [0,+,1]_1
1974 analyze_scalar_evolution_in_loop (loop1, loop3, k2) =
1975 analyze_scalar_evolution_in_loop (loop1, loop2, k2) = dont_know
1977 The value of k2 in the use in loop 1 is known, though:
1978 analyze_scalar_evolution_in_loop (loop1, loop1, k1) = [0,+,1]_1
1979 analyze_scalar_evolution_in_loop (loop1, loop1, k2) = 100
1982 static tree
1983 analyze_scalar_evolution_in_loop (struct loop *wrto_loop, struct loop *use_loop,
1984 tree version, bool *folded_casts)
1986 bool val = false;
1987 tree ev = version, tmp;
1989 /* We cannot just do
1991 tmp = analyze_scalar_evolution (use_loop, version);
1992 ev = resolve_mixers (wrto_loop, tmp);
1994 as resolve_mixers would query the scalar evolution with respect to
1995 wrto_loop. For example, in the situation described in the function
1996 comment, suppose that wrto_loop = loop1, use_loop = loop3 and
1997 version = k2. Then
1999 analyze_scalar_evolution (use_loop, version) = k2
2001 and resolve_mixers (loop1, k2) finds that the value of k2 in loop 1
2002 is 100, which is a wrong result, since we are interested in the
2003 value in loop 3.
2005 Instead, we need to proceed from use_loop to wrto_loop loop by loop,
2006 each time checking that there is no evolution in the inner loop. */
2008 if (folded_casts)
2009 *folded_casts = false;
2010 while (1)
2012 tmp = analyze_scalar_evolution (use_loop, ev);
2013 ev = resolve_mixers (use_loop, tmp);
2015 if (folded_casts && tmp != ev)
2016 *folded_casts = true;
2018 if (use_loop == wrto_loop)
2019 return ev;
2021 /* If the value of the use changes in the inner loop, we cannot express
2022 its value in the outer loop (we might try to return interval chrec,
2023 but we do not have a user for it anyway) */
2024 if (!no_evolution_in_loop_p (ev, use_loop->num, &val)
2025 || !val)
2026 return chrec_dont_know;
2028 use_loop = loop_outer (use_loop);
2032 /* Returns from CACHE the value for VERSION instantiated below
2033 INSTANTIATED_BELOW block. */
2035 static tree
2036 get_instantiated_value (htab_t cache, basic_block instantiated_below,
2037 tree version)
2039 struct scev_info_str *info, pattern;
2041 pattern.var = version;
2042 pattern.instantiated_below = instantiated_below;
2043 info = (struct scev_info_str *) htab_find (cache, &pattern);
2045 if (info)
2046 return info->chrec;
2047 else
2048 return NULL_TREE;
2051 /* Sets in CACHE the value of VERSION instantiated below basic block
2052 INSTANTIATED_BELOW to VAL. */
2054 static void
2055 set_instantiated_value (htab_t cache, basic_block instantiated_below,
2056 tree version, tree val)
2058 struct scev_info_str *info, pattern;
2059 PTR *slot;
2061 pattern.var = version;
2062 pattern.instantiated_below = instantiated_below;
2063 slot = htab_find_slot (cache, &pattern, INSERT);
2065 if (!*slot)
2066 *slot = new_scev_info_str (instantiated_below, version);
2067 info = (struct scev_info_str *) *slot;
2068 info->chrec = val;
2071 /* Return the closed_loop_phi node for VAR. If there is none, return
2072 NULL_TREE. */
2074 static tree
2075 loop_closed_phi_def (tree var)
2077 struct loop *loop;
2078 edge exit;
2079 gimple phi;
2080 gimple_stmt_iterator psi;
2082 if (var == NULL_TREE
2083 || TREE_CODE (var) != SSA_NAME)
2084 return NULL_TREE;
2086 loop = loop_containing_stmt (SSA_NAME_DEF_STMT (var));
2087 exit = single_exit (loop);
2088 if (!exit)
2089 return NULL_TREE;
2091 for (psi = gsi_start_phis (exit->dest); !gsi_end_p (psi); gsi_next (&psi))
2093 phi = gsi_stmt (psi);
2094 if (PHI_ARG_DEF_FROM_EDGE (phi, exit) == var)
2095 return PHI_RESULT (phi);
2098 return NULL_TREE;
2101 static tree instantiate_scev_r (basic_block, struct loop *, tree, bool,
2102 htab_t, int);
2104 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2105 and EVOLUTION_LOOP, that were left under a symbolic form.
2107 CHREC is an SSA_NAME to be instantiated.
2109 CACHE is the cache of already instantiated values.
2111 FOLD_CONVERSIONS should be set to true when the conversions that
2112 may wrap in signed/pointer type are folded, as long as the value of
2113 the chrec is preserved.
2115 SIZE_EXPR is used for computing the size of the expression to be
2116 instantiated, and to stop if it exceeds some limit. */
2118 static tree
2119 instantiate_scev_name (basic_block instantiate_below,
2120 struct loop *evolution_loop, tree chrec,
2121 bool fold_conversions, htab_t cache, int size_expr)
2123 tree res;
2124 struct loop *def_loop;
2125 basic_block def_bb = gimple_bb (SSA_NAME_DEF_STMT (chrec));
2127 /* A parameter (or loop invariant and we do not want to include
2128 evolutions in outer loops), nothing to do. */
2129 if (!def_bb
2130 || loop_depth (def_bb->loop_father) == 0
2131 || dominated_by_p (CDI_DOMINATORS, instantiate_below, def_bb))
2132 return chrec;
2134 /* We cache the value of instantiated variable to avoid exponential
2135 time complexity due to reevaluations. We also store the convenient
2136 value in the cache in order to prevent infinite recursion -- we do
2137 not want to instantiate the SSA_NAME if it is in a mixer
2138 structure. This is used for avoiding the instantiation of
2139 recursively defined functions, such as:
2141 | a_2 -> {0, +, 1, +, a_2}_1 */
2143 res = get_instantiated_value (cache, instantiate_below, chrec);
2144 if (res)
2145 return res;
2147 res = chrec_dont_know;
2148 set_instantiated_value (cache, instantiate_below, chrec, res);
2150 def_loop = find_common_loop (evolution_loop, def_bb->loop_father);
2152 /* If the analysis yields a parametric chrec, instantiate the
2153 result again. */
2154 res = analyze_scalar_evolution (def_loop, chrec);
2156 /* Don't instantiate loop-closed-ssa phi nodes. */
2157 if (TREE_CODE (res) == SSA_NAME
2158 && (loop_containing_stmt (SSA_NAME_DEF_STMT (res)) == NULL
2159 || (loop_depth (loop_containing_stmt (SSA_NAME_DEF_STMT (res)))
2160 > loop_depth (def_loop))))
2162 if (res == chrec)
2163 res = loop_closed_phi_def (chrec);
2164 else
2165 res = chrec;
2167 if (res == NULL_TREE
2168 || !dominated_by_p (CDI_DOMINATORS, instantiate_below,
2169 gimple_bb (SSA_NAME_DEF_STMT (res))))
2170 res = chrec_dont_know;
2173 else if (res != chrec_dont_know)
2174 res = instantiate_scev_r (instantiate_below, evolution_loop, res,
2175 fold_conversions, cache, size_expr);
2177 /* Store the correct value to the cache. */
2178 set_instantiated_value (cache, instantiate_below, chrec, res);
2179 return res;
2183 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2184 and EVOLUTION_LOOP, that were left under a symbolic form.
2186 CHREC is a polynomial chain of recurrence to be instantiated.
2188 CACHE is the cache of already instantiated values.
2190 FOLD_CONVERSIONS should be set to true when the conversions that
2191 may wrap in signed/pointer type are folded, as long as the value of
2192 the chrec is preserved.
2194 SIZE_EXPR is used for computing the size of the expression to be
2195 instantiated, and to stop if it exceeds some limit. */
2197 static tree
2198 instantiate_scev_poly (basic_block instantiate_below,
2199 struct loop *evolution_loop, tree chrec,
2200 bool fold_conversions, htab_t cache, int size_expr)
2202 tree op1;
2203 tree op0 = instantiate_scev_r (instantiate_below, evolution_loop,
2204 CHREC_LEFT (chrec), fold_conversions, cache,
2205 size_expr);
2206 if (op0 == chrec_dont_know)
2207 return chrec_dont_know;
2209 op1 = instantiate_scev_r (instantiate_below, evolution_loop,
2210 CHREC_RIGHT (chrec), fold_conversions, cache,
2211 size_expr);
2212 if (op1 == chrec_dont_know)
2213 return chrec_dont_know;
2215 if (CHREC_LEFT (chrec) != op0
2216 || CHREC_RIGHT (chrec) != op1)
2218 op1 = chrec_convert_rhs (chrec_type (op0), op1, NULL);
2219 chrec = build_polynomial_chrec (CHREC_VARIABLE (chrec), op0, op1);
2221 return chrec;
2224 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2225 and EVOLUTION_LOOP, that were left under a symbolic form.
2227 "C0 CODE C1" is a binary expression of type TYPE to be instantiated.
2229 CACHE is the cache of already instantiated values.
2231 FOLD_CONVERSIONS should be set to true when the conversions that
2232 may wrap in signed/pointer type are folded, as long as the value of
2233 the chrec is preserved.
2235 SIZE_EXPR is used for computing the size of the expression to be
2236 instantiated, and to stop if it exceeds some limit. */
2238 static tree
2239 instantiate_scev_binary (basic_block instantiate_below,
2240 struct loop *evolution_loop, tree chrec, enum tree_code code,
2241 tree type, tree c0, tree c1,
2242 bool fold_conversions, htab_t cache, int size_expr)
2244 tree op1;
2245 tree op0 = instantiate_scev_r (instantiate_below, evolution_loop,
2246 c0, fold_conversions, cache,
2247 size_expr);
2248 if (op0 == chrec_dont_know)
2249 return chrec_dont_know;
2251 op1 = instantiate_scev_r (instantiate_below, evolution_loop,
2252 c1, fold_conversions, cache,
2253 size_expr);
2254 if (op1 == chrec_dont_know)
2255 return chrec_dont_know;
2257 if (c0 != op0
2258 || c1 != op1)
2260 op0 = chrec_convert (type, op0, NULL);
2261 op1 = chrec_convert_rhs (type, op1, NULL);
2263 switch (code)
2265 case POINTER_PLUS_EXPR:
2266 case PLUS_EXPR:
2267 return chrec_fold_plus (type, op0, op1);
2269 case MINUS_EXPR:
2270 return chrec_fold_minus (type, op0, op1);
2272 case MULT_EXPR:
2273 return chrec_fold_multiply (type, op0, op1);
2275 default:
2276 gcc_unreachable ();
2280 return chrec ? chrec : fold_build2 (code, type, c0, c1);
2283 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2284 and EVOLUTION_LOOP, that were left under a symbolic form.
2286 "CHREC" that stands for a convert expression "(TYPE) OP" is to be
2287 instantiated.
2289 CACHE is the cache of already instantiated values.
2291 FOLD_CONVERSIONS should be set to true when the conversions that
2292 may wrap in signed/pointer type are folded, as long as the value of
2293 the chrec is preserved.
2295 SIZE_EXPR is used for computing the size of the expression to be
2296 instantiated, and to stop if it exceeds some limit. */
2298 static tree
2299 instantiate_scev_convert (basic_block instantiate_below,
2300 struct loop *evolution_loop, tree chrec,
2301 tree type, tree op,
2302 bool fold_conversions, htab_t cache, int size_expr)
2304 tree op0 = instantiate_scev_r (instantiate_below, evolution_loop, op,
2305 fold_conversions, cache, size_expr);
2307 if (op0 == chrec_dont_know)
2308 return chrec_dont_know;
2310 if (fold_conversions)
2312 tree tmp = chrec_convert_aggressive (type, op0);
2313 if (tmp)
2314 return tmp;
2317 if (chrec && op0 == op)
2318 return chrec;
2320 /* If we used chrec_convert_aggressive, we can no longer assume that
2321 signed chrecs do not overflow, as chrec_convert does, so avoid
2322 calling it in that case. */
2323 if (fold_conversions)
2324 return fold_convert (type, op0);
2326 return chrec_convert (type, op0, NULL);
2329 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2330 and EVOLUTION_LOOP, that were left under a symbolic form.
2332 CHREC is a BIT_NOT_EXPR or a NEGATE_EXPR expression to be instantiated.
2333 Handle ~X as -1 - X.
2334 Handle -X as -1 * X.
2336 CACHE is the cache of already instantiated values.
2338 FOLD_CONVERSIONS should be set to true when the conversions that
2339 may wrap in signed/pointer type are folded, as long as the value of
2340 the chrec is preserved.
2342 SIZE_EXPR is used for computing the size of the expression to be
2343 instantiated, and to stop if it exceeds some limit. */
2345 static tree
2346 instantiate_scev_not (basic_block instantiate_below,
2347 struct loop *evolution_loop, tree chrec,
2348 enum tree_code code, tree type, tree op,
2349 bool fold_conversions, htab_t cache, int size_expr)
2351 tree op0 = instantiate_scev_r (instantiate_below, evolution_loop, op,
2352 fold_conversions, cache, size_expr);
2354 if (op0 == chrec_dont_know)
2355 return chrec_dont_know;
2357 if (op != op0)
2359 op0 = chrec_convert (type, op0, NULL);
2361 switch (code)
2363 case BIT_NOT_EXPR:
2364 return chrec_fold_minus
2365 (type, fold_convert (type, integer_minus_one_node), op0);
2367 case NEGATE_EXPR:
2368 return chrec_fold_multiply
2369 (type, fold_convert (type, integer_minus_one_node), op0);
2371 default:
2372 gcc_unreachable ();
2376 return chrec ? chrec : fold_build1 (code, type, op0);
2379 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2380 and EVOLUTION_LOOP, that were left under a symbolic form.
2382 CHREC is an expression with 3 operands to be instantiated.
2384 CACHE is the cache of already instantiated values.
2386 FOLD_CONVERSIONS should be set to true when the conversions that
2387 may wrap in signed/pointer type are folded, as long as the value of
2388 the chrec is preserved.
2390 SIZE_EXPR is used for computing the size of the expression to be
2391 instantiated, and to stop if it exceeds some limit. */
2393 static tree
2394 instantiate_scev_3 (basic_block instantiate_below,
2395 struct loop *evolution_loop, tree chrec,
2396 bool fold_conversions, htab_t cache, int size_expr)
2398 tree op1, op2;
2399 tree op0 = instantiate_scev_r (instantiate_below, evolution_loop,
2400 TREE_OPERAND (chrec, 0),
2401 fold_conversions, cache, size_expr);
2402 if (op0 == chrec_dont_know)
2403 return chrec_dont_know;
2405 op1 = instantiate_scev_r (instantiate_below, evolution_loop,
2406 TREE_OPERAND (chrec, 1),
2407 fold_conversions, cache, size_expr);
2408 if (op1 == chrec_dont_know)
2409 return chrec_dont_know;
2411 op2 = instantiate_scev_r (instantiate_below, evolution_loop,
2412 TREE_OPERAND (chrec, 2),
2413 fold_conversions, cache, size_expr);
2414 if (op2 == chrec_dont_know)
2415 return chrec_dont_know;
2417 if (op0 == TREE_OPERAND (chrec, 0)
2418 && op1 == TREE_OPERAND (chrec, 1)
2419 && op2 == TREE_OPERAND (chrec, 2))
2420 return chrec;
2422 return fold_build3 (TREE_CODE (chrec),
2423 TREE_TYPE (chrec), op0, op1, op2);
2426 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2427 and EVOLUTION_LOOP, that were left under a symbolic form.
2429 CHREC is an expression with 2 operands to be instantiated.
2431 CACHE is the cache of already instantiated values.
2433 FOLD_CONVERSIONS should be set to true when the conversions that
2434 may wrap in signed/pointer type are folded, as long as the value of
2435 the chrec is preserved.
2437 SIZE_EXPR is used for computing the size of the expression to be
2438 instantiated, and to stop if it exceeds some limit. */
2440 static tree
2441 instantiate_scev_2 (basic_block instantiate_below,
2442 struct loop *evolution_loop, tree chrec,
2443 bool fold_conversions, htab_t cache, int size_expr)
2445 tree op1;
2446 tree op0 = instantiate_scev_r (instantiate_below, evolution_loop,
2447 TREE_OPERAND (chrec, 0),
2448 fold_conversions, cache, size_expr);
2449 if (op0 == chrec_dont_know)
2450 return chrec_dont_know;
2452 op1 = instantiate_scev_r (instantiate_below, evolution_loop,
2453 TREE_OPERAND (chrec, 1),
2454 fold_conversions, cache, size_expr);
2455 if (op1 == chrec_dont_know)
2456 return chrec_dont_know;
2458 if (op0 == TREE_OPERAND (chrec, 0)
2459 && op1 == TREE_OPERAND (chrec, 1))
2460 return chrec;
2462 return fold_build2 (TREE_CODE (chrec), TREE_TYPE (chrec), op0, op1);
2465 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2466 and EVOLUTION_LOOP, that were left under a symbolic form.
2468 CHREC is an expression with 2 operands to be instantiated.
2470 CACHE is the cache of already instantiated values.
2472 FOLD_CONVERSIONS should be set to true when the conversions that
2473 may wrap in signed/pointer type are folded, as long as the value of
2474 the chrec is preserved.
2476 SIZE_EXPR is used for computing the size of the expression to be
2477 instantiated, and to stop if it exceeds some limit. */
2479 static tree
2480 instantiate_scev_1 (basic_block instantiate_below,
2481 struct loop *evolution_loop, tree chrec,
2482 bool fold_conversions, htab_t cache, int size_expr)
2484 tree op0 = instantiate_scev_r (instantiate_below, evolution_loop,
2485 TREE_OPERAND (chrec, 0),
2486 fold_conversions, cache, size_expr);
2488 if (op0 == chrec_dont_know)
2489 return chrec_dont_know;
2491 if (op0 == TREE_OPERAND (chrec, 0))
2492 return chrec;
2494 return fold_build1 (TREE_CODE (chrec), TREE_TYPE (chrec), op0);
2497 /* Analyze all the parameters of the chrec, between INSTANTIATE_BELOW
2498 and EVOLUTION_LOOP, that were left under a symbolic form.
2500 CHREC is the scalar evolution to instantiate.
2502 CACHE is the cache of already instantiated values.
2504 FOLD_CONVERSIONS should be set to true when the conversions that
2505 may wrap in signed/pointer type are folded, as long as the value of
2506 the chrec is preserved.
2508 SIZE_EXPR is used for computing the size of the expression to be
2509 instantiated, and to stop if it exceeds some limit. */
2511 static tree
2512 instantiate_scev_r (basic_block instantiate_below,
2513 struct loop *evolution_loop, tree chrec,
2514 bool fold_conversions, htab_t cache, int size_expr)
2516 /* Give up if the expression is larger than the MAX that we allow. */
2517 if (size_expr++ > PARAM_VALUE (PARAM_SCEV_MAX_EXPR_SIZE))
2518 return chrec_dont_know;
2520 if (automatically_generated_chrec_p (chrec)
2521 || is_gimple_min_invariant (chrec))
2522 return chrec;
2524 switch (TREE_CODE (chrec))
2526 case SSA_NAME:
2527 return instantiate_scev_name (instantiate_below, evolution_loop, chrec,
2528 fold_conversions, cache, size_expr);
2530 case POLYNOMIAL_CHREC:
2531 return instantiate_scev_poly (instantiate_below, evolution_loop, chrec,
2532 fold_conversions, cache, size_expr);
2534 case POINTER_PLUS_EXPR:
2535 case PLUS_EXPR:
2536 case MINUS_EXPR:
2537 case MULT_EXPR:
2538 return instantiate_scev_binary (instantiate_below, evolution_loop, chrec,
2539 TREE_CODE (chrec), chrec_type (chrec),
2540 TREE_OPERAND (chrec, 0),
2541 TREE_OPERAND (chrec, 1),
2542 fold_conversions, cache, size_expr);
2544 CASE_CONVERT:
2545 return instantiate_scev_convert (instantiate_below, evolution_loop, chrec,
2546 TREE_TYPE (chrec), TREE_OPERAND (chrec, 0),
2547 fold_conversions, cache, size_expr);
2549 case NEGATE_EXPR:
2550 case BIT_NOT_EXPR:
2551 return instantiate_scev_not (instantiate_below, evolution_loop, chrec,
2552 TREE_CODE (chrec), TREE_TYPE (chrec),
2553 TREE_OPERAND (chrec, 0),
2554 fold_conversions, cache, size_expr);
2556 case SCEV_NOT_KNOWN:
2557 return chrec_dont_know;
2559 case SCEV_KNOWN:
2560 return chrec_known;
2562 default:
2563 break;
2566 if (VL_EXP_CLASS_P (chrec))
2567 return chrec_dont_know;
2569 switch (TREE_CODE_LENGTH (TREE_CODE (chrec)))
2571 case 3:
2572 return instantiate_scev_3 (instantiate_below, evolution_loop, chrec,
2573 fold_conversions, cache, size_expr);
2575 case 2:
2576 return instantiate_scev_2 (instantiate_below, evolution_loop, chrec,
2577 fold_conversions, cache, size_expr);
2579 case 1:
2580 return instantiate_scev_1 (instantiate_below, evolution_loop, chrec,
2581 fold_conversions, cache, size_expr);
2583 case 0:
2584 return chrec;
2586 default:
2587 break;
2590 /* Too complicated to handle. */
2591 return chrec_dont_know;
2594 /* Analyze all the parameters of the chrec that were left under a
2595 symbolic form. INSTANTIATE_BELOW is the basic block that stops the
2596 recursive instantiation of parameters: a parameter is a variable
2597 that is defined in a basic block that dominates INSTANTIATE_BELOW or
2598 a function parameter. */
2600 tree
2601 instantiate_scev (basic_block instantiate_below, struct loop *evolution_loop,
2602 tree chrec)
2604 tree res;
2605 htab_t cache = htab_create (10, hash_scev_info, eq_scev_info, del_scev_info);
2607 if (dump_file && (dump_flags & TDF_DETAILS))
2609 fprintf (dump_file, "(instantiate_scev \n");
2610 fprintf (dump_file, " (instantiate_below = %d)\n", instantiate_below->index);
2611 fprintf (dump_file, " (evolution_loop = %d)\n", evolution_loop->num);
2612 fprintf (dump_file, " (chrec = ");
2613 print_generic_expr (dump_file, chrec, 0);
2614 fprintf (dump_file, ")\n");
2617 res = instantiate_scev_r (instantiate_below, evolution_loop, chrec, false,
2618 cache, 0);
2620 if (dump_file && (dump_flags & TDF_DETAILS))
2622 fprintf (dump_file, " (res = ");
2623 print_generic_expr (dump_file, res, 0);
2624 fprintf (dump_file, "))\n");
2627 htab_delete (cache);
2629 return res;
2632 /* Similar to instantiate_parameters, but does not introduce the
2633 evolutions in outer loops for LOOP invariants in CHREC, and does not
2634 care about causing overflows, as long as they do not affect value
2635 of an expression. */
2637 tree
2638 resolve_mixers (struct loop *loop, tree chrec)
2640 htab_t cache = htab_create (10, hash_scev_info, eq_scev_info, del_scev_info);
2641 tree ret = instantiate_scev_r (block_before_loop (loop), loop, chrec, true,
2642 cache, 0);
2643 htab_delete (cache);
2644 return ret;
2647 /* Entry point for the analysis of the number of iterations pass.
2648 This function tries to safely approximate the number of iterations
2649 the loop will run. When this property is not decidable at compile
2650 time, the result is chrec_dont_know. Otherwise the result is
2651 a scalar or a symbolic parameter.
2653 Example of analysis: suppose that the loop has an exit condition:
2655 "if (b > 49) goto end_loop;"
2657 and that in a previous analysis we have determined that the
2658 variable 'b' has an evolution function:
2660 "EF = {23, +, 5}_2".
2662 When we evaluate the function at the point 5, i.e. the value of the
2663 variable 'b' after 5 iterations in the loop, we have EF (5) = 48,
2664 and EF (6) = 53. In this case the value of 'b' on exit is '53' and
2665 the loop body has been executed 6 times. */
2667 tree
2668 number_of_latch_executions (struct loop *loop)
2670 tree res, type;
2671 edge exit;
2672 struct tree_niter_desc niter_desc;
2674 /* Determine whether the number_of_iterations_in_loop has already
2675 been computed. */
2676 res = loop->nb_iterations;
2677 if (res)
2678 return res;
2679 res = chrec_dont_know;
2681 if (dump_file && (dump_flags & TDF_DETAILS))
2682 fprintf (dump_file, "(number_of_iterations_in_loop\n");
2684 exit = single_exit (loop);
2685 if (!exit)
2686 goto end;
2688 if (!number_of_iterations_exit (loop, exit, &niter_desc, false))
2689 goto end;
2691 type = TREE_TYPE (niter_desc.niter);
2692 if (integer_nonzerop (niter_desc.may_be_zero))
2693 res = build_int_cst (type, 0);
2694 else if (integer_zerop (niter_desc.may_be_zero))
2695 res = niter_desc.niter;
2696 else
2697 res = chrec_dont_know;
2699 end:
2700 return set_nb_iterations_in_loop (loop, res);
2703 /* Returns the number of executions of the exit condition of LOOP,
2704 i.e., the number by one higher than number_of_latch_executions.
2705 Note that unlike number_of_latch_executions, this number does
2706 not necessarily fit in the unsigned variant of the type of
2707 the control variable -- if the number of iterations is a constant,
2708 we return chrec_dont_know if adding one to number_of_latch_executions
2709 overflows; however, in case the number of iterations is symbolic
2710 expression, the caller is responsible for dealing with this
2711 the possible overflow. */
2713 tree
2714 number_of_exit_cond_executions (struct loop *loop)
2716 tree ret = number_of_latch_executions (loop);
2717 tree type = chrec_type (ret);
2719 if (chrec_contains_undetermined (ret))
2720 return ret;
2722 ret = chrec_fold_plus (type, ret, build_int_cst (type, 1));
2723 if (TREE_CODE (ret) == INTEGER_CST
2724 && TREE_OVERFLOW (ret))
2725 return chrec_dont_know;
2727 return ret;
2730 /* One of the drivers for testing the scalar evolutions analysis.
2731 This function computes the number of iterations for all the loops
2732 from the EXIT_CONDITIONS array. */
2734 static void
2735 number_of_iterations_for_all_loops (VEC(gimple,heap) **exit_conditions)
2737 unsigned int i;
2738 unsigned nb_chrec_dont_know_loops = 0;
2739 unsigned nb_static_loops = 0;
2740 gimple cond;
2742 for (i = 0; VEC_iterate (gimple, *exit_conditions, i, cond); i++)
2744 tree res = number_of_latch_executions (loop_containing_stmt (cond));
2745 if (chrec_contains_undetermined (res))
2746 nb_chrec_dont_know_loops++;
2747 else
2748 nb_static_loops++;
2751 if (dump_file)
2753 fprintf (dump_file, "\n(\n");
2754 fprintf (dump_file, "-----------------------------------------\n");
2755 fprintf (dump_file, "%d\tnb_chrec_dont_know_loops\n", nb_chrec_dont_know_loops);
2756 fprintf (dump_file, "%d\tnb_static_loops\n", nb_static_loops);
2757 fprintf (dump_file, "%d\tnb_total_loops\n", number_of_loops ());
2758 fprintf (dump_file, "-----------------------------------------\n");
2759 fprintf (dump_file, ")\n\n");
2761 print_loops (dump_file, 3);
2767 /* Counters for the stats. */
2769 struct chrec_stats
2771 unsigned nb_chrecs;
2772 unsigned nb_affine;
2773 unsigned nb_affine_multivar;
2774 unsigned nb_higher_poly;
2775 unsigned nb_chrec_dont_know;
2776 unsigned nb_undetermined;
2779 /* Reset the counters. */
2781 static inline void
2782 reset_chrecs_counters (struct chrec_stats *stats)
2784 stats->nb_chrecs = 0;
2785 stats->nb_affine = 0;
2786 stats->nb_affine_multivar = 0;
2787 stats->nb_higher_poly = 0;
2788 stats->nb_chrec_dont_know = 0;
2789 stats->nb_undetermined = 0;
2792 /* Dump the contents of a CHREC_STATS structure. */
2794 static void
2795 dump_chrecs_stats (FILE *file, struct chrec_stats *stats)
2797 fprintf (file, "\n(\n");
2798 fprintf (file, "-----------------------------------------\n");
2799 fprintf (file, "%d\taffine univariate chrecs\n", stats->nb_affine);
2800 fprintf (file, "%d\taffine multivariate chrecs\n", stats->nb_affine_multivar);
2801 fprintf (file, "%d\tdegree greater than 2 polynomials\n",
2802 stats->nb_higher_poly);
2803 fprintf (file, "%d\tchrec_dont_know chrecs\n", stats->nb_chrec_dont_know);
2804 fprintf (file, "-----------------------------------------\n");
2805 fprintf (file, "%d\ttotal chrecs\n", stats->nb_chrecs);
2806 fprintf (file, "%d\twith undetermined coefficients\n",
2807 stats->nb_undetermined);
2808 fprintf (file, "-----------------------------------------\n");
2809 fprintf (file, "%d\tchrecs in the scev database\n",
2810 (int) htab_elements (scalar_evolution_info));
2811 fprintf (file, "%d\tsets in the scev database\n", nb_set_scev);
2812 fprintf (file, "%d\tgets in the scev database\n", nb_get_scev);
2813 fprintf (file, "-----------------------------------------\n");
2814 fprintf (file, ")\n\n");
2817 /* Gather statistics about CHREC. */
2819 static void
2820 gather_chrec_stats (tree chrec, struct chrec_stats *stats)
2822 if (dump_file && (dump_flags & TDF_STATS))
2824 fprintf (dump_file, "(classify_chrec ");
2825 print_generic_expr (dump_file, chrec, 0);
2826 fprintf (dump_file, "\n");
2829 stats->nb_chrecs++;
2831 if (chrec == NULL_TREE)
2833 stats->nb_undetermined++;
2834 return;
2837 switch (TREE_CODE (chrec))
2839 case POLYNOMIAL_CHREC:
2840 if (evolution_function_is_affine_p (chrec))
2842 if (dump_file && (dump_flags & TDF_STATS))
2843 fprintf (dump_file, " affine_univariate\n");
2844 stats->nb_affine++;
2846 else if (evolution_function_is_affine_multivariate_p (chrec, 0))
2848 if (dump_file && (dump_flags & TDF_STATS))
2849 fprintf (dump_file, " affine_multivariate\n");
2850 stats->nb_affine_multivar++;
2852 else
2854 if (dump_file && (dump_flags & TDF_STATS))
2855 fprintf (dump_file, " higher_degree_polynomial\n");
2856 stats->nb_higher_poly++;
2859 break;
2861 default:
2862 break;
2865 if (chrec_contains_undetermined (chrec))
2867 if (dump_file && (dump_flags & TDF_STATS))
2868 fprintf (dump_file, " undetermined\n");
2869 stats->nb_undetermined++;
2872 if (dump_file && (dump_flags & TDF_STATS))
2873 fprintf (dump_file, ")\n");
2876 /* One of the drivers for testing the scalar evolutions analysis.
2877 This function analyzes the scalar evolution of all the scalars
2878 defined as loop phi nodes in one of the loops from the
2879 EXIT_CONDITIONS array.
2881 TODO Optimization: A loop is in canonical form if it contains only
2882 a single scalar loop phi node. All the other scalars that have an
2883 evolution in the loop are rewritten in function of this single
2884 index. This allows the parallelization of the loop. */
2886 static void
2887 analyze_scalar_evolution_for_all_loop_phi_nodes (VEC(gimple,heap) **exit_conditions)
2889 unsigned int i;
2890 struct chrec_stats stats;
2891 gimple cond, phi;
2892 gimple_stmt_iterator psi;
2894 reset_chrecs_counters (&stats);
2896 for (i = 0; VEC_iterate (gimple, *exit_conditions, i, cond); i++)
2898 struct loop *loop;
2899 basic_block bb;
2900 tree chrec;
2902 loop = loop_containing_stmt (cond);
2903 bb = loop->header;
2905 for (psi = gsi_start_phis (bb); !gsi_end_p (psi); gsi_next (&psi))
2907 phi = gsi_stmt (psi);
2908 if (is_gimple_reg (PHI_RESULT (phi)))
2910 chrec = instantiate_parameters
2911 (loop,
2912 analyze_scalar_evolution (loop, PHI_RESULT (phi)));
2914 if (dump_file && (dump_flags & TDF_STATS))
2915 gather_chrec_stats (chrec, &stats);
2920 if (dump_file && (dump_flags & TDF_STATS))
2921 dump_chrecs_stats (dump_file, &stats);
2924 /* Callback for htab_traverse, gathers information on chrecs in the
2925 hashtable. */
2927 static int
2928 gather_stats_on_scev_database_1 (void **slot, void *stats)
2930 struct scev_info_str *entry = (struct scev_info_str *) *slot;
2932 gather_chrec_stats (entry->chrec, (struct chrec_stats *) stats);
2934 return 1;
2937 /* Classify the chrecs of the whole database. */
2939 void
2940 gather_stats_on_scev_database (void)
2942 struct chrec_stats stats;
2944 if (!dump_file)
2945 return;
2947 reset_chrecs_counters (&stats);
2949 htab_traverse (scalar_evolution_info, gather_stats_on_scev_database_1,
2950 &stats);
2952 dump_chrecs_stats (dump_file, &stats);
2957 /* Initializer. */
2959 static void
2960 initialize_scalar_evolutions_analyzer (void)
2962 /* The elements below are unique. */
2963 if (chrec_dont_know == NULL_TREE)
2965 chrec_not_analyzed_yet = NULL_TREE;
2966 chrec_dont_know = make_node (SCEV_NOT_KNOWN);
2967 chrec_known = make_node (SCEV_KNOWN);
2968 TREE_TYPE (chrec_dont_know) = void_type_node;
2969 TREE_TYPE (chrec_known) = void_type_node;
2973 /* Initialize the analysis of scalar evolutions for LOOPS. */
2975 void
2976 scev_initialize (void)
2978 loop_iterator li;
2979 struct loop *loop;
2981 scalar_evolution_info = htab_create_alloc (100,
2982 hash_scev_info,
2983 eq_scev_info,
2984 del_scev_info,
2985 ggc_calloc,
2986 ggc_free);
2988 initialize_scalar_evolutions_analyzer ();
2990 FOR_EACH_LOOP (li, loop, 0)
2992 loop->nb_iterations = NULL_TREE;
2996 /* Cleans up the information cached by the scalar evolutions analysis. */
2998 void
2999 scev_reset (void)
3001 loop_iterator li;
3002 struct loop *loop;
3004 if (!scalar_evolution_info || !current_loops)
3005 return;
3007 htab_empty (scalar_evolution_info);
3008 FOR_EACH_LOOP (li, loop, 0)
3010 loop->nb_iterations = NULL_TREE;
3014 /* Checks whether use of OP in USE_LOOP behaves as a simple affine iv with
3015 respect to WRTO_LOOP and returns its base and step in IV if possible
3016 (see analyze_scalar_evolution_in_loop for more details on USE_LOOP
3017 and WRTO_LOOP). If ALLOW_NONCONSTANT_STEP is true, we want step to be
3018 invariant in LOOP. Otherwise we require it to be an integer constant.
3020 IV->no_overflow is set to true if we are sure the iv cannot overflow (e.g.
3021 because it is computed in signed arithmetics). Consequently, adding an
3022 induction variable
3024 for (i = IV->base; ; i += IV->step)
3026 is only safe if IV->no_overflow is false, or TYPE_OVERFLOW_UNDEFINED is
3027 false for the type of the induction variable, or you can prove that i does
3028 not wrap by some other argument. Otherwise, this might introduce undefined
3029 behavior, and
3031 for (i = iv->base; ; i = (type) ((unsigned type) i + (unsigned type) iv->step))
3033 must be used instead. */
3035 bool
3036 simple_iv (struct loop *wrto_loop, struct loop *use_loop, tree op,
3037 affine_iv *iv, bool allow_nonconstant_step)
3039 tree type, ev;
3040 bool folded_casts;
3042 iv->base = NULL_TREE;
3043 iv->step = NULL_TREE;
3044 iv->no_overflow = false;
3046 type = TREE_TYPE (op);
3047 if (TREE_CODE (type) != INTEGER_TYPE
3048 && TREE_CODE (type) != POINTER_TYPE)
3049 return false;
3051 ev = analyze_scalar_evolution_in_loop (wrto_loop, use_loop, op,
3052 &folded_casts);
3053 if (chrec_contains_undetermined (ev)
3054 || chrec_contains_symbols_defined_in_loop (ev, wrto_loop->num))
3055 return false;
3057 if (tree_does_not_contain_chrecs (ev))
3059 iv->base = ev;
3060 iv->step = build_int_cst (TREE_TYPE (ev), 0);
3061 iv->no_overflow = true;
3062 return true;
3065 if (TREE_CODE (ev) != POLYNOMIAL_CHREC
3066 || CHREC_VARIABLE (ev) != (unsigned) wrto_loop->num)
3067 return false;
3069 iv->step = CHREC_RIGHT (ev);
3070 if ((!allow_nonconstant_step && TREE_CODE (iv->step) != INTEGER_CST)
3071 || tree_contains_chrecs (iv->step, NULL))
3072 return false;
3074 iv->base = CHREC_LEFT (ev);
3075 if (tree_contains_chrecs (iv->base, NULL))
3076 return false;
3078 iv->no_overflow = !folded_casts && TYPE_OVERFLOW_UNDEFINED (type);
3080 return true;
3083 /* Runs the analysis of scalar evolutions. */
3085 void
3086 scev_analysis (void)
3088 VEC(gimple,heap) *exit_conditions;
3090 exit_conditions = VEC_alloc (gimple, heap, 37);
3091 select_loops_exit_conditions (&exit_conditions);
3093 if (dump_file && (dump_flags & TDF_STATS))
3094 analyze_scalar_evolution_for_all_loop_phi_nodes (&exit_conditions);
3096 number_of_iterations_for_all_loops (&exit_conditions);
3097 VEC_free (gimple, heap, exit_conditions);
3100 /* Finalize the scalar evolution analysis. */
3102 void
3103 scev_finalize (void)
3105 if (!scalar_evolution_info)
3106 return;
3107 htab_delete (scalar_evolution_info);
3108 scalar_evolution_info = NULL;
3111 /* Returns true if the expression EXPR is considered to be too expensive
3112 for scev_const_prop. */
3114 bool
3115 expression_expensive_p (tree expr)
3117 enum tree_code code;
3119 if (is_gimple_val (expr))
3120 return false;
3122 code = TREE_CODE (expr);
3123 if (code == TRUNC_DIV_EXPR
3124 || code == CEIL_DIV_EXPR
3125 || code == FLOOR_DIV_EXPR
3126 || code == ROUND_DIV_EXPR
3127 || code == TRUNC_MOD_EXPR
3128 || code == CEIL_MOD_EXPR
3129 || code == FLOOR_MOD_EXPR
3130 || code == ROUND_MOD_EXPR
3131 || code == EXACT_DIV_EXPR)
3133 /* Division by power of two is usually cheap, so we allow it.
3134 Forbid anything else. */
3135 if (!integer_pow2p (TREE_OPERAND (expr, 1)))
3136 return true;
3139 switch (TREE_CODE_CLASS (code))
3141 case tcc_binary:
3142 case tcc_comparison:
3143 if (expression_expensive_p (TREE_OPERAND (expr, 1)))
3144 return true;
3146 /* Fallthru. */
3147 case tcc_unary:
3148 return expression_expensive_p (TREE_OPERAND (expr, 0));
3150 default:
3151 return true;
3155 /* Replace ssa names for that scev can prove they are constant by the
3156 appropriate constants. Also perform final value replacement in loops,
3157 in case the replacement expressions are cheap.
3159 We only consider SSA names defined by phi nodes; rest is left to the
3160 ordinary constant propagation pass. */
3162 unsigned int
3163 scev_const_prop (void)
3165 basic_block bb;
3166 tree name, type, ev;
3167 gimple phi, ass;
3168 struct loop *loop, *ex_loop;
3169 bitmap ssa_names_to_remove = NULL;
3170 unsigned i;
3171 loop_iterator li;
3172 gimple_stmt_iterator psi;
3174 if (number_of_loops () <= 1)
3175 return 0;
3177 FOR_EACH_BB (bb)
3179 loop = bb->loop_father;
3181 for (psi = gsi_start_phis (bb); !gsi_end_p (psi); gsi_next (&psi))
3183 phi = gsi_stmt (psi);
3184 name = PHI_RESULT (phi);
3186 if (!is_gimple_reg (name))
3187 continue;
3189 type = TREE_TYPE (name);
3191 if (!POINTER_TYPE_P (type)
3192 && !INTEGRAL_TYPE_P (type))
3193 continue;
3195 ev = resolve_mixers (loop, analyze_scalar_evolution (loop, name));
3196 if (!is_gimple_min_invariant (ev)
3197 || !may_propagate_copy (name, ev))
3198 continue;
3200 /* Replace the uses of the name. */
3201 if (name != ev)
3202 replace_uses_by (name, ev);
3204 if (!ssa_names_to_remove)
3205 ssa_names_to_remove = BITMAP_ALLOC (NULL);
3206 bitmap_set_bit (ssa_names_to_remove, SSA_NAME_VERSION (name));
3210 /* Remove the ssa names that were replaced by constants. We do not
3211 remove them directly in the previous cycle, since this
3212 invalidates scev cache. */
3213 if (ssa_names_to_remove)
3215 bitmap_iterator bi;
3217 EXECUTE_IF_SET_IN_BITMAP (ssa_names_to_remove, 0, i, bi)
3219 gimple_stmt_iterator psi;
3220 name = ssa_name (i);
3221 phi = SSA_NAME_DEF_STMT (name);
3223 gcc_assert (gimple_code (phi) == GIMPLE_PHI);
3224 psi = gsi_for_stmt (phi);
3225 remove_phi_node (&psi, true);
3228 BITMAP_FREE (ssa_names_to_remove);
3229 scev_reset ();
3232 /* Now the regular final value replacement. */
3233 FOR_EACH_LOOP (li, loop, LI_FROM_INNERMOST)
3235 edge exit;
3236 tree def, rslt, niter;
3237 gimple_stmt_iterator bsi;
3239 /* If we do not know exact number of iterations of the loop, we cannot
3240 replace the final value. */
3241 exit = single_exit (loop);
3242 if (!exit)
3243 continue;
3245 niter = number_of_latch_executions (loop);
3246 if (niter == chrec_dont_know)
3247 continue;
3249 /* Ensure that it is possible to insert new statements somewhere. */
3250 if (!single_pred_p (exit->dest))
3251 split_loop_exit_edge (exit);
3252 bsi = gsi_after_labels (exit->dest);
3254 ex_loop = superloop_at_depth (loop,
3255 loop_depth (exit->dest->loop_father) + 1);
3257 for (psi = gsi_start_phis (exit->dest); !gsi_end_p (psi); )
3259 phi = gsi_stmt (psi);
3260 rslt = PHI_RESULT (phi);
3261 def = PHI_ARG_DEF_FROM_EDGE (phi, exit);
3262 if (!is_gimple_reg (def))
3264 gsi_next (&psi);
3265 continue;
3268 if (!POINTER_TYPE_P (TREE_TYPE (def))
3269 && !INTEGRAL_TYPE_P (TREE_TYPE (def)))
3271 gsi_next (&psi);
3272 continue;
3275 def = analyze_scalar_evolution_in_loop (ex_loop, loop, def, NULL);
3276 def = compute_overall_effect_of_inner_loop (ex_loop, def);
3277 if (!tree_does_not_contain_chrecs (def)
3278 || chrec_contains_symbols_defined_in_loop (def, ex_loop->num)
3279 /* Moving the computation from the loop may prolong life range
3280 of some ssa names, which may cause problems if they appear
3281 on abnormal edges. */
3282 || contains_abnormal_ssa_name_p (def)
3283 /* Do not emit expensive expressions. The rationale is that
3284 when someone writes a code like
3286 while (n > 45) n -= 45;
3288 he probably knows that n is not large, and does not want it
3289 to be turned into n %= 45. */
3290 || expression_expensive_p (def))
3292 gsi_next (&psi);
3293 continue;
3296 /* Eliminate the PHI node and replace it by a computation outside
3297 the loop. */
3298 def = unshare_expr (def);
3299 remove_phi_node (&psi, false);
3301 def = force_gimple_operand_gsi (&bsi, def, false, NULL_TREE,
3302 true, GSI_SAME_STMT);
3303 ass = gimple_build_assign (rslt, def);
3304 gsi_insert_before (&bsi, ass, GSI_SAME_STMT);
3307 return 0;
3310 #include "gt-tree-scalar-evolution.h"