2 * Copyright (C) 1995-2008 University of Karlsruhe. All right reserved.
4 * This file is part of libFirm.
6 * This file may be distributed and/or modified under the terms of the
7 * GNU General Public License version 2 as published by the Free Software
8 * Foundation and appearing in the file LICENSE.GPL included in the
9 * packaging of this file.
11 * Licensees holding valid libFirm Professional Edition licenses may use
12 * this file in accordance with the libFirm Commercial License.
13 * Agreement provided with the Software.
15 * This file is provided AS IS with NO WARRANTY OF ANY KIND, INCLUDING THE
16 * WARRANTY OF DESIGN, MERCHANTABILITY AND FITNESS FOR A PARTICULAR
22 * @brief Dead node elimination and Procedure Inlining.
23 * @author Michael Beck, Goetz Lindenmaier
32 #include "irgraph_t.h"
35 #include "iroptimize.h"
52 #include "irbackedge_t.h"
53 #include "opt_inline_t.h"
58 #include "analyze_irg_args.h"
59 #include "iredges_t.h"
63 #include "iropt_dbg.h"
66 DEBUG_ONLY(static firm_dbg_module_t *dbg;)
68 /*------------------------------------------------------------------*/
69 /* Routines for dead node elimination / copying garbage collection */
71 /*------------------------------------------------------------------*/
74 * Remember the new node in the old node by using a field all nodes have.
76 #define set_new_node(oldn, newn) set_irn_link(oldn, newn)
79 * Get this new node, before the old node is forgotten.
81 #define get_new_node(oldn) get_irn_link(oldn)
84 * Check if a new node was set.
86 #define has_new_node(n) (get_new_node(n) != NULL)
89 * We use the block_visited flag to mark that we have computed the
90 * number of useful predecessors for this block.
91 * Further we encode the new arity in this flag in the old blocks.
92 * Remembering the arity is useful, as it saves a lot of pointer
93 * accesses. This function is called for all Phi and Block nodes
97 compute_new_arity(ir_node *b) {
98 int i, res, irn_arity;
101 irg_v = get_irg_block_visited(current_ir_graph);
102 block_v = get_Block_block_visited(b);
103 if (block_v >= irg_v) {
104 /* we computed the number of preds for this block and saved it in the
106 return block_v - irg_v;
108 /* compute the number of good predecessors */
109 res = irn_arity = get_irn_arity(b);
110 for (i = 0; i < irn_arity; i++)
111 if (is_Bad(get_irn_n(b, i))) res--;
112 /* save it in the flag. */
113 set_Block_block_visited(b, irg_v + res);
119 * Copies the node to the new obstack. The Ins of the new node point to
120 * the predecessors on the old obstack. For block/phi nodes not all
121 * predecessors might be copied. n->link points to the new node.
122 * For Phi and Block nodes the function allocates in-arrays with an arity
123 * only for useful predecessors. The arity is determined by counting
124 * the non-bad predecessors of the block.
126 * @param n The node to be copied
127 * @param env if non-NULL, the node number attribute will be copied to the new node
129 * Note: Also used for loop unrolling.
131 static void copy_node(ir_node *n, void *env) {
134 ir_op *op = get_irn_op(n);
138 /* node copied already */
140 } else if (op == op_Block) {
142 new_arity = compute_new_arity(n);
143 n->attr.block.graph_arr = NULL;
145 block = get_nodes_block(n);
147 new_arity = compute_new_arity(block);
149 new_arity = get_irn_arity(n);
152 nn = new_ir_node(get_irn_dbg_info(n),
159 /* Copy the attributes. These might point to additional data. If this
160 was allocated on the old obstack the pointers now are dangling. This
161 frees e.g. the memory of the graph_arr allocated in new_immBlock. */
162 if (op == op_Block) {
163 /* we cannot allow blocks WITHOUT macroblock input */
164 set_Block_MacroBlock(nn, get_Block_MacroBlock(n));
166 copy_node_attr(n, nn);
169 /* for easier debugging, we want to copy the node numbers too */
170 nn->node_nr = n->node_nr;
174 hook_dead_node_elim_subst(current_ir_graph, n, nn);
178 * Copies new predecessors of old node to new node remembered in link.
179 * Spare the Bad predecessors of Phi and Block nodes.
181 static void copy_preds(ir_node *n, void *env) {
186 nn = get_new_node(n);
189 /* copy the macro block header */
190 ir_node *mbh = get_Block_MacroBlock(n);
193 /* this block is a macroblock header */
194 set_Block_MacroBlock(nn, nn);
196 /* get the macro block header */
197 ir_node *nmbh = get_new_node(mbh);
198 assert(nmbh != NULL);
199 set_Block_MacroBlock(nn, nmbh);
202 /* Don't copy Bad nodes. */
204 irn_arity = get_irn_arity(n);
205 for (i = 0; i < irn_arity; i++) {
206 if (! is_Bad(get_irn_n(n, i))) {
207 ir_node *pred = get_irn_n(n, i);
208 set_irn_n(nn, j, get_new_node(pred));
212 /* repair the block visited flag from above misuse. Repair it in both
213 graphs so that the old one can still be used. */
214 set_Block_block_visited(nn, 0);
215 set_Block_block_visited(n, 0);
216 /* Local optimization could not merge two subsequent blocks if
217 in array contained Bads. Now it's possible.
218 We don't call optimize_in_place as it requires
219 that the fields in ir_graph are set properly. */
220 if (!has_Block_entity(nn) &&
221 get_opt_control_flow_straightening() &&
222 get_Block_n_cfgpreds(nn) == 1 &&
223 is_Jmp(get_Block_cfgpred(nn, 0))) {
224 ir_node *old = get_nodes_block(get_Block_cfgpred(nn, 0));
226 /* Jmp jumps into the block it is in -- deal self cycle. */
227 assert(is_Bad(get_new_node(get_irg_bad(current_ir_graph))));
228 exchange(nn, get_new_node(get_irg_bad(current_ir_graph)));
233 } else if (is_Phi(n) && get_irn_arity(n) > 0) {
234 /* Don't copy node if corresponding predecessor in block is Bad.
235 The Block itself should not be Bad. */
236 block = get_nodes_block(n);
237 set_nodes_block(nn, get_new_node(block));
239 irn_arity = get_irn_arity(n);
240 for (i = 0; i < irn_arity; i++) {
241 if (! is_Bad(get_irn_n(block, i))) {
242 ir_node *pred = get_irn_n(n, i);
243 set_irn_n(nn, j, get_new_node(pred));
244 /*if (is_backedge(n, i)) set_backedge(nn, j);*/
248 /* If the pre walker reached this Phi after the post walker visited the
249 block block_visited is > 0. */
250 set_Block_block_visited(get_nodes_block(n), 0);
251 /* Compacting the Phi's ins might generate Phis with only one
253 if (get_irn_arity(nn) == 1)
254 exchange(nn, get_irn_n(nn, 0));
256 irn_arity = get_irn_arity(n);
257 for (i = -1; i < irn_arity; i++)
258 set_irn_n(nn, i, get_new_node(get_irn_n(n, i)));
260 /* Now the new node is complete. We can add it to the hash table for CSE.
261 @@@ inlining aborts if we identify End. Why? */
263 add_identities(current_ir_graph->value_table, nn);
267 * Copies the graph recursively, compacts the keep-alives of the end node.
269 * @param irg the graph to be copied
270 * @param copy_node_nr If non-zero, the node number will be copied
272 static void copy_graph(ir_graph *irg, int copy_node_nr) {
273 ir_node *oe, *ne, *ob, *nb, *om, *nm; /* old end, new end, old bad, new bad, old NoMem, new NoMem */
274 ir_node *ka; /* keep alive */
278 /* Some nodes must be copied by hand, sigh */
279 vfl = get_irg_visited(irg);
280 set_irg_visited(irg, vfl + 1);
282 oe = get_irg_end(irg);
283 mark_irn_visited(oe);
284 /* copy the end node by hand, allocate dynamic in array! */
285 ne = new_ir_node(get_irn_dbg_info(oe),
292 /* Copy the attributes. Well, there might be some in the future... */
293 copy_node_attr(oe, ne);
294 set_new_node(oe, ne);
296 /* copy the Bad node */
297 ob = get_irg_bad(irg);
298 mark_irn_visited(ob);
299 nb = new_ir_node(get_irn_dbg_info(ob),
306 copy_node_attr(ob, nb);
307 set_new_node(ob, nb);
309 /* copy the NoMem node */
310 om = get_irg_no_mem(irg);
311 mark_irn_visited(om);
312 nm = new_ir_node(get_irn_dbg_info(om),
319 copy_node_attr(om, nm);
320 set_new_node(om, nm);
322 /* copy the live nodes */
323 set_irg_visited(irg, vfl);
324 irg_walk(get_nodes_block(oe), copy_node, copy_preds, INT_TO_PTR(copy_node_nr));
326 /* Note: from yet, the visited flag of the graph is equal to vfl + 1 */
328 /* visit the anchors as well */
329 for (i = get_irg_n_anchors(irg) - 1; i >= 0; --i) {
330 ir_node *n = get_irg_anchor(irg, i);
332 if (n && (get_irn_visited(n) <= vfl)) {
333 set_irg_visited(irg, vfl);
334 irg_walk(n, copy_node, copy_preds, INT_TO_PTR(copy_node_nr));
338 /* copy_preds for the end node ... */
339 set_nodes_block(ne, get_new_node(get_nodes_block(oe)));
341 /*- ... and now the keep alives. -*/
342 /* First pick the not marked block nodes and walk them. We must pick these
343 first as else we will oversee blocks reachable from Phis. */
344 irn_arity = get_End_n_keepalives(oe);
345 for (i = 0; i < irn_arity; i++) {
346 ka = get_End_keepalive(oe, i);
348 if (get_irn_visited(ka) <= vfl) {
349 /* We must keep the block alive and copy everything reachable */
350 set_irg_visited(irg, vfl);
351 irg_walk(ka, copy_node, copy_preds, INT_TO_PTR(copy_node_nr));
353 add_End_keepalive(ne, get_new_node(ka));
357 /* Now pick other nodes. Here we will keep all! */
358 irn_arity = get_End_n_keepalives(oe);
359 for (i = 0; i < irn_arity; i++) {
360 ka = get_End_keepalive(oe, i);
362 if (get_irn_visited(ka) <= vfl) {
363 /* We didn't copy the node yet. */
364 set_irg_visited(irg, vfl);
365 irg_walk(ka, copy_node, copy_preds, INT_TO_PTR(copy_node_nr));
367 add_End_keepalive(ne, get_new_node(ka));
371 /* start block sometimes only reached after keep alives */
372 set_nodes_block(nb, get_new_node(get_nodes_block(ob)));
373 set_nodes_block(nm, get_new_node(get_nodes_block(om)));
377 * Copies the graph reachable from current_ir_graph->end to the obstack
378 * in current_ir_graph and fixes the environment.
379 * Then fixes the fields in current_ir_graph containing nodes of the
382 * @param copy_node_nr If non-zero, the node number will be copied
385 copy_graph_env(int copy_node_nr) {
386 ir_graph *irg = current_ir_graph;
387 ir_node *old_end, *new_anchor;
390 /* remove end_except and end_reg nodes */
391 old_end = get_irg_end(irg);
392 set_irg_end_except (irg, old_end);
393 set_irg_end_reg (irg, old_end);
395 /* Not all nodes remembered in irg might be reachable
396 from the end node. Assure their link is set to NULL, so that
397 we can test whether new nodes have been computed. */
398 for (i = get_irg_n_anchors(irg) - 1; i >= 0; --i) {
399 ir_node *n = get_irg_anchor(irg, i);
401 set_new_node(n, NULL);
403 /* we use the block walk flag for removing Bads from Blocks ins. */
404 inc_irg_block_visited(irg);
407 copy_graph(irg, copy_node_nr);
410 old_end = get_irg_end(irg);
411 new_anchor = new_Anchor(irg);
413 for (i = get_irg_n_anchors(irg) - 1; i >= 0; --i) {
414 ir_node *n = get_irg_anchor(irg, i);
416 set_irn_n(new_anchor, i, get_new_node(n));
419 irg->anchor = new_anchor;
421 /* ensure the new anchor is placed in the endblock */
422 set_nodes_block(new_anchor, get_irg_end_block(irg));
426 * Copies all reachable nodes to a new obstack. Removes bad inputs
427 * from block nodes and the corresponding inputs from Phi nodes.
428 * Merges single exit blocks with single entry blocks and removes
430 * Adds all new nodes to a new hash table for CSE. Does not
431 * perform CSE, so the hash table might contain common subexpressions.
433 void dead_node_elimination(ir_graph *irg) {
435 #ifdef INTERPROCEDURAL_VIEW
436 int rem_ipview = get_interprocedural_view();
438 struct obstack *graveyard_obst = NULL;
439 struct obstack *rebirth_obst = NULL;
441 edges_deactivate(irg);
443 /* inform statistics that we started a dead-node elimination run */
444 hook_dead_node_elim(irg, 1);
446 /* Remember external state of current_ir_graph. */
447 rem = current_ir_graph;
448 current_ir_graph = irg;
449 #ifdef INTERPROCEDURAL_VIEW
450 set_interprocedural_view(0);
453 assert(get_irg_phase_state(irg) != phase_building);
455 /* Handle graph state */
456 free_callee_info(irg);
460 /* @@@ so far we loose loops when copying */
461 free_loop_information(irg);
463 set_irg_doms_inconsistent(irg);
465 /* A quiet place, where the old obstack can rest in peace,
466 until it will be cremated. */
467 graveyard_obst = irg->obst;
469 /* A new obstack, where the reachable nodes will be copied to. */
470 rebirth_obst = XMALLOC(struct obstack);
471 irg->obst = rebirth_obst;
472 obstack_init(irg->obst);
473 irg->last_node_idx = 0;
475 /* We also need a new value table for CSE */
476 del_identities(irg->value_table);
477 irg->value_table = new_identities();
479 /* Copy the graph from the old to the new obstack */
480 copy_graph_env(/*copy_node_nr=*/1);
482 /* Free memory from old unoptimized obstack */
483 obstack_free(graveyard_obst, 0); /* First empty the obstack ... */
484 xfree(graveyard_obst); /* ... then free it. */
486 /* inform statistics that the run is over */
487 hook_dead_node_elim(irg, 0);
489 current_ir_graph = rem;
490 #ifdef INTERPROCEDURAL_VIEW
491 set_interprocedural_view(rem_ipview);
495 ir_graph_pass_t *dead_node_elimination_pass(const char *name) {
496 return def_graph_pass(name ? name : "dce", dead_node_elimination);
500 * Relink bad predecessors of a block and store the old in array to the
501 * link field. This function is called by relink_bad_predecessors().
502 * The array of link field starts with the block operand at position 0.
503 * If block has bad predecessors, create a new in array without bad preds.
504 * Otherwise let in array untouched.
506 static void relink_bad_block_predecessors(ir_node *n, void *env) {
507 ir_node **new_in, *irn;
508 int i, new_irn_n, old_irn_arity, new_irn_arity = 0;
511 /* if link field of block is NULL, look for bad predecessors otherwise
512 this is already done */
513 if (is_Block(n) && get_irn_link(n) == NULL) {
514 /* save old predecessors in link field (position 0 is the block operand)*/
515 set_irn_link(n, get_irn_in(n));
517 /* count predecessors without bad nodes */
518 old_irn_arity = get_irn_arity(n);
519 for (i = 0; i < old_irn_arity; i++)
520 if (!is_Bad(get_irn_n(n, i)))
523 /* arity changing: set new predecessors without bad nodes */
524 if (new_irn_arity < old_irn_arity) {
525 /* Get new predecessor array. We do not resize the array, as we must
526 keep the old one to update Phis. */
527 new_in = NEW_ARR_D(ir_node *, current_ir_graph->obst, (new_irn_arity+1));
529 /* set new predecessors in array */
532 for (i = 0; i < old_irn_arity; i++) {
533 irn = get_irn_n(n, i);
535 new_in[new_irn_n] = irn;
536 is_backedge(n, i) ? set_backedge(n, new_irn_n-1) : set_not_backedge(n, new_irn_n-1);
540 /* ARR_SETLEN(int, n->attr.block.backedge, new_irn_arity); */
541 ARR_SHRINKLEN(n->attr.block.backedge, new_irn_arity);
543 } /* ir node has bad predecessors */
544 } /* Block is not relinked */
548 * Relinks Bad predecessors from Blocks and Phis called by walker
549 * remove_bad_predecesors(). If n is a Block, call
550 * relink_bad_block_redecessors(). If n is a Phi-node, call also the relinking
551 * function of Phi's Block. If this block has bad predecessors, relink preds
554 static void relink_bad_predecessors(ir_node *n, void *env) {
555 ir_node *block, **old_in;
556 int i, old_irn_arity, new_irn_arity;
558 /* relink bad predecessors of a block */
560 relink_bad_block_predecessors(n, env);
562 /* If Phi node relink its block and its predecessors */
564 /* Relink predecessors of phi's block */
565 block = get_nodes_block(n);
566 if (get_irn_link(block) == NULL)
567 relink_bad_block_predecessors(block, env);
569 old_in = (ir_node **)get_irn_link(block); /* Of Phi's Block */
570 old_irn_arity = ARR_LEN(old_in);
572 /* Relink Phi predecessors if count of predecessors changed */
573 if (old_irn_arity != ARR_LEN(get_irn_in(block))) {
574 /* set new predecessors in array
575 n->in[0] remains the same block */
577 for(i = 1; i < old_irn_arity; i++)
578 if (!is_Bad(old_in[i])) {
579 n->in[new_irn_arity] = n->in[i];
580 is_backedge(n, i) ? set_backedge(n, new_irn_arity) : set_not_backedge(n, new_irn_arity);
584 ARR_SETLEN(ir_node *, n->in, new_irn_arity);
585 ARR_SETLEN(int, n->attr.phi.u.backedge, new_irn_arity);
587 } /* n is a Phi node */
591 * Removes Bad Bad predecessors from Blocks and the corresponding
592 * inputs to Phi nodes as in dead_node_elimination but without
594 * On walking up set the link field to NULL, on walking down call
595 * relink_bad_predecessors() (This function stores the old in array
596 * to the link field and sets a new in array if arity of predecessors
599 void remove_bad_predecessors(ir_graph *irg) {
600 panic("Fix backedge handling first");
601 irg_walk_graph(irg, firm_clear_link, relink_bad_predecessors, NULL);
608 __)|_| | \_/ | \_/(/_ |_/\__|__
610 The following stuff implements a facility that automatically patches
611 registered ir_node pointers to the new node when a dead node elimination occurs.
614 struct _survive_dce_t {
618 hook_entry_t dead_node_elim;
619 hook_entry_t dead_node_elim_subst;
622 typedef struct _survive_dce_list_t {
623 struct _survive_dce_list_t *next;
625 } survive_dce_list_t;
627 static void dead_node_hook(void *context, ir_graph *irg, int start) {
628 survive_dce_t *sd = context;
631 /* Create a new map before the dead node elimination is performed. */
633 sd->new_places = pmap_create_ex(pmap_count(sd->places));
635 /* Patch back all nodes if dead node elimination is over and something is to be done. */
636 pmap_destroy(sd->places);
637 sd->places = sd->new_places;
638 sd->new_places = NULL;
643 * Hook called when dead node elimination replaces old by nw.
645 static void dead_node_subst_hook(void *context, ir_graph *irg, ir_node *old, ir_node *nw) {
646 survive_dce_t *sd = context;
647 survive_dce_list_t *list = pmap_get(sd->places, old);
650 /* If the node is to be patched back, write the new address to all registered locations. */
652 survive_dce_list_t *p;
654 for (p = list; p; p = p->next)
657 pmap_insert(sd->new_places, nw, list);
662 * Make a new Survive DCE environment.
664 survive_dce_t *new_survive_dce(void) {
665 survive_dce_t *res = XMALLOC(survive_dce_t);
666 obstack_init(&res->obst);
667 res->places = pmap_create();
668 res->new_places = NULL;
670 res->dead_node_elim.hook._hook_dead_node_elim = dead_node_hook;
671 res->dead_node_elim.context = res;
672 res->dead_node_elim.next = NULL;
674 res->dead_node_elim_subst.hook._hook_dead_node_elim_subst = dead_node_subst_hook;
675 res->dead_node_elim_subst.context = res;
676 res->dead_node_elim_subst.next = NULL;
678 register_hook(hook_dead_node_elim, &res->dead_node_elim);
679 register_hook(hook_dead_node_elim_subst, &res->dead_node_elim_subst);
684 * Free a Survive DCE environment.
686 void free_survive_dce(survive_dce_t *sd) {
687 obstack_free(&sd->obst, NULL);
688 pmap_destroy(sd->places);
689 unregister_hook(hook_dead_node_elim, &sd->dead_node_elim);
690 unregister_hook(hook_dead_node_elim_subst, &sd->dead_node_elim_subst);
695 * Register a node pointer to be patched upon DCE.
696 * When DCE occurs, the node pointer specified by @p place will be
697 * patched to the new address of the node it is pointing to.
699 * @param sd The Survive DCE environment.
700 * @param place The address of the node pointer.
702 void survive_dce_register_irn(survive_dce_t *sd, ir_node **place) {
703 if (*place != NULL) {
704 ir_node *irn = *place;
705 survive_dce_list_t *curr = pmap_get(sd->places, irn);
706 survive_dce_list_t *nw = OALLOC(&sd->obst, survive_dce_list_t);
711 pmap_insert(sd->places, irn, nw);
715 /*--------------------------------------------------------------------*/
716 /* Functionality for inlining */
717 /*--------------------------------------------------------------------*/
720 * Copy node for inlineing. Updates attributes that change when
721 * inlineing but not for dead node elimination.
723 * Copies the node by calling copy_node() and then updates the entity if
724 * it's a local one. env must be a pointer of the frame type of the
725 * inlined procedure. The new entities must be in the link field of
728 static void copy_node_inline(ir_node *n, void *env) {
730 ir_type *frame_tp = (ir_type *)env;
734 nn = get_new_node(n);
736 /* use copied entities from the new frame */
737 if (get_entity_owner(get_Sel_entity(n)) == frame_tp) {
738 set_Sel_entity(nn, get_entity_link(get_Sel_entity(n)));
740 } else if (is_Block(n)) {
741 nn = get_new_node(n);
742 nn->attr.block.irg.irg = current_ir_graph;
747 * Copies new predecessors of old node and move constants to
750 static void copy_preds_inline(ir_node *n, void *env) {
754 nn = skip_Id(get_new_node(n));
755 if (is_irn_constlike(nn)) {
756 /* move Constants into the start block */
757 set_nodes_block(nn, get_irg_start_block(current_ir_graph));
759 n = identify_remember(current_ir_graph->value_table, nn);
768 * Walker: checks if P_value_arg_base is used.
770 static void find_addr(ir_node *node, void *env) {
771 int *allow_inline = env;
773 ir_graph *irg = current_ir_graph;
774 if (get_Sel_ptr(node) == get_irg_frame(irg)) {
775 /* access to frame */
776 ir_entity *ent = get_Sel_entity(node);
777 if (get_entity_owner(ent) != get_irg_frame_type(irg)) {
778 /* access to value_type */
782 } else if (is_Alloc(node) && get_Alloc_where(node) == stack_alloc) {
784 * Refuse to inline alloca call unless user explicitly forced so as this
785 * may change program's memory overhead drastically when the function
786 * using alloca is called in loop. In GCC present in SPEC2000 inlining
787 * into schedule_block cause it to require 2GB of ram instead of 256MB.
789 * Sorrily this is true with our implementation also.
790 * Moreover, we cannot differentiate between alloca() and VLA yet, so this
791 * disables inlining of functions using VLA (with are completely save).
794 * - add a flag to the Alloc node for "real" alloca() calls
795 * - add a new Stack-Restore node at the end of a function using alloca()
802 * Check if we can inline a given call.
803 * Currently, we cannot inline two cases:
804 * - call with compound arguments
805 * - graphs that take the address of a parameter
807 * check these conditions here
809 static int can_inline(ir_node *call, ir_graph *called_graph) {
810 ir_type *call_type = get_Call_type(call);
811 int params, ress, i, res;
812 assert(is_Method_type(call_type));
814 params = get_method_n_params(call_type);
815 ress = get_method_n_ress(call_type);
817 /* check parameters for compound arguments */
818 for (i = 0; i < params; ++i) {
819 ir_type *p_type = get_method_param_type(call_type, i);
821 if (is_compound_type(p_type))
825 /* check results for compound arguments */
826 for (i = 0; i < ress; ++i) {
827 ir_type *r_type = get_method_res_type(call_type, i);
829 if (is_compound_type(r_type))
834 irg_walk_graph(called_graph, find_addr, NULL, &res);
840 exc_handler, /**< There is a handler. */
841 exc_no_handler /**< Exception handling not represented. */
844 /* Inlines a method at the given call site. */
845 int inline_method(ir_node *call, ir_graph *called_graph) {
847 ir_node *post_call, *post_bl;
848 ir_node *in[pn_Start_max];
849 ir_node *end, *end_bl, *block;
854 int arity, n_ret, n_exc, n_res, i, n, j, rem_opt, irn_arity, n_params;
856 enum exc_mode exc_handling;
857 ir_type *called_frame, *curr_frame, *mtp, *ctp;
860 irg_inline_property prop = get_irg_inline_property(called_graph);
861 unsigned long visited;
863 if (prop == irg_inline_forbidden)
866 ent = get_irg_entity(called_graph);
868 mtp = get_entity_type(ent);
869 ctp = get_Call_type(call);
870 n_params = get_method_n_params(mtp);
871 n_res = get_method_n_ress(mtp);
872 if (n_params > get_method_n_params(ctp)) {
873 /* this is a bad feature of C: without a prototype, we can
874 * call a function with less parameters than needed. Currently
875 * we don't support this, although we could use Unknown than. */
878 if (n_res != get_method_n_ress(ctp)) {
882 /* Argh, compiling C has some bad consequences:
883 * It is implementation dependent what happens in that case.
884 * We support inlining, if the bitsize of the types matches AND
885 * the same arithmetic is used. */
886 for (i = n_params - 1; i >= 0; --i) {
887 ir_type *param_tp = get_method_param_type(mtp, i);
888 ir_type *arg_tp = get_method_param_type(ctp, i);
890 if (param_tp != arg_tp) {
891 ir_mode *pmode = get_type_mode(param_tp);
892 ir_mode *amode = get_type_mode(arg_tp);
894 if (pmode == NULL || amode == NULL)
896 if (get_mode_size_bits(pmode) != get_mode_size_bits(amode))
898 if (get_mode_arithmetic(pmode) != get_mode_arithmetic(amode))
900 /* otherwise we can simply "reinterpret" the bits */
903 for (i = n_res - 1; i >= 0; --i) {
904 ir_type *decl_res_tp = get_method_res_type(mtp, i);
905 ir_type *used_res_tp = get_method_res_type(ctp, i);
907 if (decl_res_tp != used_res_tp) {
908 ir_mode *decl_mode = get_type_mode(decl_res_tp);
909 ir_mode *used_mode = get_type_mode(used_res_tp);
910 if (decl_mode == NULL || used_mode == NULL)
912 if (get_mode_size_bits(decl_mode) != get_mode_size_bits(used_mode))
914 if (get_mode_arithmetic(decl_mode) != get_mode_arithmetic(used_mode))
916 /* otherwise we can "reinterpret" the bits */
920 irg = get_irn_irg(call);
923 * We cannot inline a recursive call. The graph must be copied before
924 * the call the inline_method() using create_irg_copy().
926 if (called_graph == irg)
930 * currently, we cannot inline two cases:
931 * - call with compound arguments
932 * - graphs that take the address of a parameter
934 if (! can_inline(call, called_graph))
937 rem = current_ir_graph;
938 current_ir_graph = irg;
940 DB((dbg, LEVEL_1, "Inlining %+F(%+F) into %+F\n", call, called_graph, irg));
942 /* -- Turn off optimizations, this can cause problems when allocating new nodes. -- */
943 rem_opt = get_opt_optimize();
946 /* Handle graph state */
947 assert(get_irg_phase_state(irg) != phase_building);
948 assert(get_irg_pinned(irg) == op_pin_state_pinned);
949 assert(get_irg_pinned(called_graph) == op_pin_state_pinned);
950 set_irg_outs_inconsistent(irg);
951 set_irg_extblk_inconsistent(irg);
952 set_irg_doms_inconsistent(irg);
953 set_irg_loopinfo_inconsistent(irg);
954 set_irg_callee_info_state(irg, irg_callee_info_inconsistent);
955 set_irg_entity_usage_state(irg, ir_entity_usage_not_computed);
957 /* -- Check preconditions -- */
958 assert(is_Call(call));
960 /* here we know we WILL inline, so inform the statistics */
961 hook_inline(call, called_graph);
963 /* -- Decide how to handle exception control flow: Is there a handler
964 for the Call node, or do we branch directly to End on an exception?
966 0 There is a handler.
967 2 Exception handling not represented in Firm. -- */
969 ir_node *Xproj = NULL;
971 for (proj = get_irn_link(call); proj; proj = get_irn_link(proj)) {
972 long proj_nr = get_Proj_proj(proj);
973 if (proj_nr == pn_Call_X_except) Xproj = proj;
975 exc_handling = Xproj != NULL ? exc_handler : exc_no_handler;
978 /* create the argument tuple */
979 NEW_ARR_A(ir_type *, args_in, n_params);
981 block = get_nodes_block(call);
982 for (i = n_params - 1; i >= 0; --i) {
983 ir_node *arg = get_Call_param(call, i);
984 ir_type *param_tp = get_method_param_type(mtp, i);
985 ir_mode *mode = get_type_mode(param_tp);
987 if (mode != get_irn_mode(arg)) {
988 arg = new_r_Conv(block, arg, mode);
994 the procedure and later replaces the Start node of the called graph.
995 Post_call is the old Call node and collects the results of the called
996 graph. Both will end up being a tuple. -- */
997 post_bl = get_nodes_block(call);
998 set_irg_current_block(irg, post_bl);
999 /* XxMxPxPxPxT of Start + parameter of Call */
1000 in[pn_Start_X_initial_exec] = new_Jmp();
1001 in[pn_Start_M] = get_Call_mem(call);
1002 in[pn_Start_P_frame_base] = get_irg_frame(irg);
1003 in[pn_Start_P_tls] = get_irg_tls(irg);
1004 in[pn_Start_T_args] = new_Tuple(n_params, args_in);
1005 pre_call = new_Tuple(pn_Start_max, in);
1009 The new block gets the ins of the old block, pre_call and all its
1010 predecessors and all Phi nodes. -- */
1011 part_block(pre_call);
1013 /* -- Prepare state for dead node elimination -- */
1014 /* Visited flags in calling irg must be >= flag in called irg.
1015 Else walker and arity computation will not work. */
1016 if (get_irg_visited(irg) <= get_irg_visited(called_graph))
1017 set_irg_visited(irg, get_irg_visited(called_graph) + 1);
1018 if (get_irg_block_visited(irg) < get_irg_block_visited(called_graph))
1019 set_irg_block_visited(irg, get_irg_block_visited(called_graph));
1020 visited = get_irg_visited(irg);
1022 /* Set pre_call as new Start node in link field of the start node of
1023 calling graph and pre_calls block as new block for the start block
1025 Further mark these nodes so that they are not visited by the
1027 set_irn_link(get_irg_start(called_graph), pre_call);
1028 set_irn_visited(get_irg_start(called_graph), visited);
1029 set_irn_link(get_irg_start_block(called_graph), get_nodes_block(pre_call));
1030 set_irn_visited(get_irg_start_block(called_graph), visited);
1032 set_irn_link(get_irg_bad(called_graph), get_irg_bad(current_ir_graph));
1033 set_irn_visited(get_irg_bad(called_graph), visited);
1035 set_irn_link(get_irg_no_mem(called_graph), get_irg_no_mem(current_ir_graph));
1036 set_irn_visited(get_irg_no_mem(called_graph), visited);
1038 /* Initialize for compaction of in arrays */
1039 inc_irg_block_visited(irg);
1041 /* -- Replicate local entities of the called_graph -- */
1042 /* copy the entities. */
1043 irp_reserve_resources(irp, IR_RESOURCE_ENTITY_LINK);
1044 called_frame = get_irg_frame_type(called_graph);
1045 curr_frame = get_irg_frame_type(irg);
1046 for (i = 0, n = get_class_n_members(called_frame); i < n; ++i) {
1047 ir_entity *new_ent, *old_ent;
1048 old_ent = get_class_member(called_frame, i);
1049 new_ent = copy_entity_own(old_ent, curr_frame);
1050 set_entity_link(old_ent, new_ent);
1053 /* visited is > than that of called graph. With this trick visited will
1054 remain unchanged so that an outer walker, e.g., searching the call nodes
1055 to inline, calling this inline will not visit the inlined nodes. */
1056 set_irg_visited(irg, get_irg_visited(irg)-1);
1058 /* -- Performing dead node elimination inlines the graph -- */
1059 /* Copies the nodes to the obstack of current_ir_graph. Updates links to new
1061 irg_walk(get_irg_end(called_graph), copy_node_inline, copy_preds_inline,
1062 get_irg_frame_type(called_graph));
1064 irp_free_resources(irp, IR_RESOURCE_ENTITY_LINK);
1066 /* Repair called_graph */
1067 set_irg_visited(called_graph, get_irg_visited(irg));
1068 set_irg_block_visited(called_graph, get_irg_block_visited(irg));
1069 set_Block_block_visited(get_irg_start_block(called_graph), 0);
1071 /* -- Merge the end of the inlined procedure with the call site -- */
1072 /* We will turn the old Call node into a Tuple with the following
1075 0: Phi of all Memories of Return statements.
1076 1: Jmp from new Block that merges the control flow from all exception
1077 predecessors of the old end block.
1078 2: Tuple of all arguments.
1079 3: Phi of Exception memories.
1080 In case the old Call directly branches to End on an exception we don't
1081 need the block merging all exceptions nor the Phi of the exception
1085 /* -- Precompute some values -- */
1086 end_bl = get_new_node(get_irg_end_block(called_graph));
1087 end = get_new_node(get_irg_end(called_graph));
1088 arity = get_Block_n_cfgpreds(end_bl); /* arity = n_exc + n_ret */
1089 n_res = get_method_n_ress(get_Call_type(call));
1091 res_pred = XMALLOCN(ir_node*, n_res);
1092 cf_pred = XMALLOCN(ir_node*, arity);
1094 set_irg_current_block(irg, post_bl); /* just to make sure */
1096 /* -- archive keepalives -- */
1097 irn_arity = get_irn_arity(end);
1098 for (i = 0; i < irn_arity; i++) {
1099 ir_node *ka = get_End_keepalive(end, i);
1101 add_End_keepalive(get_irg_end(irg), ka);
1104 /* The new end node will die. We need not free as the in array is on the obstack:
1105 copy_node() only generated 'D' arrays. */
1107 /* -- Replace Return nodes by Jump nodes. -- */
1109 for (i = 0; i < arity; i++) {
1111 ret = get_Block_cfgpred(end_bl, i);
1112 if (is_Return(ret)) {
1113 cf_pred[n_ret] = new_r_Jmp(get_nodes_block(ret));
1117 set_irn_in(post_bl, n_ret, cf_pred);
1119 /* -- Build a Tuple for all results of the method.
1120 Add Phi node if there was more than one Return. -- */
1121 turn_into_tuple(post_call, pn_Call_max);
1122 /* First the Memory-Phi */
1124 for (i = 0; i < arity; i++) {
1125 ret = get_Block_cfgpred(end_bl, i);
1126 if (is_Return(ret)) {
1127 cf_pred[n_mem_phi++] = get_Return_mem(ret);
1129 /* memory output for some exceptions is directly connected to End */
1131 cf_pred[n_mem_phi++] = new_r_Proj(get_nodes_block(ret), ret, mode_M, 3);
1132 } else if (is_fragile_op(ret)) {
1133 /* We rely that all cfops have the memory output at the same position. */
1134 cf_pred[n_mem_phi++] = new_r_Proj(get_nodes_block(ret), ret, mode_M, 0);
1135 } else if (is_Raise(ret)) {
1136 cf_pred[n_mem_phi++] = new_r_Proj(get_nodes_block(ret), ret, mode_M, 1);
1139 phi = new_Phi(n_mem_phi, cf_pred, mode_M);
1140 set_Tuple_pred(call, pn_Call_M, phi);
1141 /* Conserve Phi-list for further inlinings -- but might be optimized */
1142 if (get_nodes_block(phi) == post_bl) {
1143 set_irn_link(phi, get_irn_link(post_bl));
1144 set_irn_link(post_bl, phi);
1146 /* Now the real results */
1148 for (j = 0; j < n_res; j++) {
1149 ir_type *res_type = get_method_res_type(ctp, j);
1150 ir_mode *res_mode = get_type_mode(res_type);
1152 for (i = 0; i < arity; i++) {
1153 ret = get_Block_cfgpred(end_bl, i);
1154 if (is_Return(ret)) {
1155 ir_node *res = get_Return_res(ret, j);
1156 if (get_irn_mode(res) != res_mode) {
1157 ir_node *block = get_nodes_block(res);
1158 res = new_r_Conv(block, res, res_mode);
1160 cf_pred[n_ret] = res;
1165 phi = new_Phi(n_ret, cf_pred, get_irn_mode(cf_pred[0]));
1169 /* Conserve Phi-list for further inlinings -- but might be optimized */
1170 if (get_nodes_block(phi) == post_bl) {
1171 set_Phi_next(phi, get_Block_phis(post_bl));
1172 set_Block_phis(post_bl, phi);
1175 set_Tuple_pred(call, pn_Call_T_result, new_Tuple(n_res, res_pred));
1177 set_Tuple_pred(call, pn_Call_T_result, new_Bad());
1179 /* handle the regular call */
1180 set_Tuple_pred(call, pn_Call_X_regular, new_Jmp());
1182 /* For now, we cannot inline calls with value_base */
1183 set_Tuple_pred(call, pn_Call_P_value_res_base, new_Bad());
1185 /* Finally the exception control flow.
1186 We have two possible situations:
1187 First if the Call branches to an exception handler:
1188 We need to add a Phi node to
1189 collect the memory containing the exception objects. Further we need
1190 to add another block to get a correct representation of this Phi. To
1191 this block we add a Jmp that resolves into the X output of the Call
1192 when the Call is turned into a tuple.
1193 Second: There is no exception edge. Just add all inlined exception
1194 branches to the End node.
1196 if (exc_handling == exc_handler) {
1198 for (i = 0; i < arity; i++) {
1200 ret = get_Block_cfgpred(end_bl, i);
1201 irn = skip_Proj(ret);
1202 if (is_fragile_op(irn) || is_Raise(irn)) {
1203 cf_pred[n_exc] = ret;
1208 ir_node *block = new_Block(n_exc, cf_pred);
1209 set_cur_block(block);
1210 set_Tuple_pred(call, pn_Call_X_except, new_Jmp());
1212 set_Tuple_pred(call, pn_Call_X_except, new_Bad());
1215 ir_node *main_end_bl;
1216 int main_end_bl_arity;
1217 ir_node **end_preds;
1219 /* assert(exc_handling == 1 || no exceptions. ) */
1221 for (i = 0; i < arity; i++) {
1222 ir_node *ret = get_Block_cfgpred(end_bl, i);
1223 ir_node *irn = skip_Proj(ret);
1225 if (is_fragile_op(irn) || is_Raise(irn)) {
1226 cf_pred[n_exc] = ret;
1230 main_end_bl = get_irg_end_block(irg);
1231 main_end_bl_arity = get_irn_arity(main_end_bl);
1232 end_preds = XMALLOCN(ir_node*, n_exc + main_end_bl_arity);
1234 for (i = 0; i < main_end_bl_arity; ++i)
1235 end_preds[i] = get_irn_n(main_end_bl, i);
1236 for (i = 0; i < n_exc; ++i)
1237 end_preds[main_end_bl_arity + i] = cf_pred[i];
1238 set_irn_in(main_end_bl, n_exc + main_end_bl_arity, end_preds);
1239 set_Tuple_pred(call, pn_Call_X_except, new_Bad());
1245 /* -- Turn CSE back on. -- */
1246 set_optimize(rem_opt);
1247 current_ir_graph = rem;
1252 /********************************************************************/
1253 /* Apply inlining to small methods. */
1254 /********************************************************************/
1256 static struct obstack temp_obst;
1258 /** Represents a possible inlinable call in a graph. */
1259 typedef struct _call_entry {
1260 ir_node *call; /**< The Call node. */
1261 ir_graph *callee; /**< The callee IR-graph. */
1262 list_head list; /**< List head for linking the next one. */
1263 int loop_depth; /**< The loop depth of this call. */
1264 int benefice; /**< The calculated benefice of this call. */
1265 unsigned local_adr:1; /**< Set if this call gets an address of a local variable. */
1266 unsigned all_const:1; /**< Set if this call has only constant parameters. */
1270 * environment for inlining small irgs
1272 typedef struct _inline_env_t {
1273 struct obstack obst; /**< An obstack where call_entries are allocated on. */
1274 list_head calls; /**< The call entry list. */
1278 * Returns the irg called from a Call node. If the irg is not
1279 * known, NULL is returned.
1281 * @param call the call node
1283 static ir_graph *get_call_called_irg(ir_node *call) {
1286 addr = get_Call_ptr(call);
1287 if (is_Global(addr)) {
1288 ir_entity *ent = get_Global_entity(addr);
1289 return get_entity_irg(ent);
1296 * Walker: Collect all calls to known graphs inside a graph.
1298 static void collect_calls(ir_node *call, void *env) {
1300 if (is_Call(call)) {
1301 ir_graph *called_irg = get_call_called_irg(call);
1303 if (called_irg != NULL) {
1304 /* The Call node calls a locally defined method. Remember to inline. */
1305 inline_env_t *ienv = env;
1306 call_entry *entry = OALLOC(&ienv->obst, call_entry);
1308 entry->callee = called_irg;
1309 entry->loop_depth = 0;
1310 entry->benefice = 0;
1311 entry->local_adr = 0;
1312 entry->all_const = 0;
1314 list_add_tail(&entry->list, &ienv->calls);
1320 * Inlines all small methods at call sites where the called address comes
1321 * from a Const node that references the entity representing the called
1323 * The size argument is a rough measure for the code size of the method:
1324 * Methods where the obstack containing the firm graph is smaller than
1327 void inline_small_irgs(ir_graph *irg, int size) {
1328 ir_graph *rem = current_ir_graph;
1332 current_ir_graph = irg;
1333 /* Handle graph state */
1334 assert(get_irg_phase_state(irg) != phase_building);
1335 free_callee_info(irg);
1337 /* Find Call nodes to inline.
1338 (We can not inline during a walk of the graph, as inlining the same
1339 method several times changes the visited flag of the walked graph:
1340 after the first inlining visited of the callee equals visited of
1341 the caller. With the next inlining both are increased.) */
1342 obstack_init(&env.obst);
1343 INIT_LIST_HEAD(&env.calls);
1344 irg_walk_graph(irg, NULL, collect_calls, &env);
1346 if (! list_empty(&env.calls)) {
1347 /* There are calls to inline */
1348 ir_reserve_resources(irg, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
1349 collect_phiprojs(irg);
1351 list_for_each_entry(call_entry, entry, &env.calls, list) {
1352 ir_graph *callee = entry->callee;
1353 irg_inline_property prop = get_irg_inline_property(callee);
1355 if (prop == irg_inline_forbidden || get_irg_additional_properties(callee) & mtp_property_weak) {
1356 /* do not inline forbidden / weak graphs */
1360 if (prop >= irg_inline_forced ||
1361 _obstack_memory_used(callee->obst) - (int)obstack_room(callee->obst) < size) {
1362 inline_method(entry->call, callee);
1365 ir_free_resources(irg, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
1367 obstack_free(&env.obst, NULL);
1368 current_ir_graph = rem;
1371 struct inline_small_irgs_pass_t {
1372 ir_graph_pass_t pass;
1377 * Wrapper to run inline_small_irgs() as a pass.
1379 static int inline_small_irgs_wrapper(ir_graph *irg, void *context) {
1380 struct inline_small_irgs_pass_t *pass = context;
1382 inline_small_irgs(irg, pass->size);
1386 /* create a pass for inline_small_irgs() */
1387 ir_graph_pass_t *inline_small_irgs_pass(const char *name, int size) {
1388 struct inline_small_irgs_pass_t *pass =
1389 XMALLOCZ(struct inline_small_irgs_pass_t);
1392 return def_graph_pass_constructor(
1393 &pass->pass, name ? name : "inline_small_irgs", inline_small_irgs_wrapper);
1397 * Environment for inlining irgs.
1400 list_head calls; /**< List of of all call nodes in this graph. */
1401 unsigned *local_weights; /**< Once allocated, the beneficial weight for transmitting local addresses. */
1402 unsigned n_nodes; /**< Number of nodes in graph except Id, Tuple, Proj, Start, End. */
1403 unsigned n_blocks; /**< Number of Blocks in graph without Start and End block. */
1404 unsigned n_nodes_orig; /**< for statistics */
1405 unsigned n_call_nodes; /**< Number of Call nodes in the graph. */
1406 unsigned n_call_nodes_orig; /**< for statistics */
1407 unsigned n_callers; /**< Number of known graphs that call this graphs. */
1408 unsigned n_callers_orig; /**< for statistics */
1409 unsigned got_inline:1; /**< Set, if at least one call inside this graph was inlined. */
1410 unsigned local_vars:1; /**< Set, if an inlined function got the address of a local variable. */
1411 unsigned recursive:1; /**< Set, if this function is self recursive. */
1415 * Allocate a new environment for inlining.
1417 static inline_irg_env *alloc_inline_irg_env(void) {
1418 inline_irg_env *env = OALLOC(&temp_obst, inline_irg_env);
1419 INIT_LIST_HEAD(&env->calls);
1420 env->local_weights = NULL;
1421 env->n_nodes = -2; /* do not count count Start, End */
1422 env->n_blocks = -2; /* do not count count Start, End Block */
1423 env->n_nodes_orig = -2; /* do not count Start, End */
1424 env->n_call_nodes = 0;
1425 env->n_call_nodes_orig = 0;
1427 env->n_callers_orig = 0;
1428 env->got_inline = 0;
1429 env->local_vars = 0;
1434 typedef struct walker_env {
1435 inline_irg_env *x; /**< the inline environment */
1436 char ignore_runtime; /**< the ignore runtime flag */
1437 char ignore_callers; /**< if set, do change callers data */
1441 * post-walker: collect all calls in the inline-environment
1442 * of a graph and sum some statistics.
1444 static void collect_calls2(ir_node *call, void *ctx) {
1446 inline_irg_env *x = env->x;
1447 ir_opcode code = get_irn_opcode(call);
1451 /* count meaningful nodes in irg */
1452 if (code != iro_Proj && code != iro_Tuple && code != iro_Sync) {
1453 if (code != iro_Block) {
1461 if (code != iro_Call) return;
1463 /* check, if it's a runtime call */
1464 if (env->ignore_runtime) {
1465 ir_node *symc = get_Call_ptr(call);
1467 if (is_Global(symc)) {
1468 ir_entity *ent = get_Global_entity(symc);
1470 if (get_entity_additional_properties(ent) & mtp_property_runtime)
1475 /* collect all call nodes */
1477 ++x->n_call_nodes_orig;
1479 callee = get_call_called_irg(call);
1480 if (callee != NULL) {
1481 if (! env->ignore_callers) {
1482 inline_irg_env *callee_env = get_irg_link(callee);
1483 /* count all static callers */
1484 ++callee_env->n_callers;
1485 ++callee_env->n_callers_orig;
1487 if (callee == current_ir_graph)
1490 /* link it in the list of possible inlinable entries */
1491 entry = OALLOC(&temp_obst, call_entry);
1493 entry->callee = callee;
1494 entry->loop_depth = get_irn_loop(get_nodes_block(call))->depth;
1495 entry->benefice = 0;
1496 entry->local_adr = 0;
1497 entry->all_const = 0;
1499 list_add_tail(&entry->list, &x->calls);
1504 * Returns TRUE if the number of callers is 0 in the irg's environment,
1505 * hence this irg is a leave.
1507 inline static int is_leave(ir_graph *irg) {
1508 inline_irg_env *env = get_irg_link(irg);
1509 return env->n_call_nodes == 0;
1513 * Returns TRUE if the number of nodes in the callee is
1514 * smaller then size in the irg's environment.
1516 inline static int is_smaller(ir_graph *callee, unsigned size) {
1517 inline_irg_env *env = get_irg_link(callee);
1518 return env->n_nodes < size;
1522 * Duplicate a call entry.
1524 * @param entry the original entry to duplicate
1525 * @param new_call the new call node
1526 * @param loop_depth_delta
1527 * delta value for the loop depth
1529 static call_entry *duplicate_call_entry(const call_entry *entry,
1530 ir_node *new_call, int loop_depth_delta) {
1531 call_entry *nentry = OALLOC(&temp_obst, call_entry);
1532 nentry->call = new_call;
1533 nentry->callee = entry->callee;
1534 nentry->benefice = entry->benefice;
1535 nentry->loop_depth = entry->loop_depth + loop_depth_delta;
1536 nentry->local_adr = entry->local_adr;
1537 nentry->all_const = entry->all_const;
1543 * Append all call nodes of the source environment to the nodes of in the destination
1546 * @param dst destination environment
1547 * @param src source environment
1548 * @param loop_depth the loop depth of the call that is replaced by the src list
1550 static void append_call_list(inline_irg_env *dst, inline_irg_env *src, int loop_depth) {
1551 call_entry *entry, *nentry;
1553 /* Note that the src list points to Call nodes in the inlined graph, but
1554 we need Call nodes in our graph. Luckily the inliner leaves this information
1555 in the link field. */
1556 list_for_each_entry(call_entry, entry, &src->calls, list) {
1557 nentry = duplicate_call_entry(entry, get_irn_link(entry->call), loop_depth);
1558 list_add_tail(&nentry->list, &dst->calls);
1560 dst->n_call_nodes += src->n_call_nodes;
1561 dst->n_nodes += src->n_nodes;
1565 * Inlines small leave methods at call sites where the called address comes
1566 * from a Const node that references the entity representing the called
1568 * The size argument is a rough measure for the code size of the method:
1569 * Methods where the obstack containing the firm graph is smaller than
1572 void inline_leave_functions(unsigned maxsize, unsigned leavesize,
1573 unsigned size, int ignore_runtime)
1575 inline_irg_env *env;
1581 call_entry *entry, *next;
1582 const call_entry *centry;
1583 pmap *copied_graphs;
1584 pmap_entry *pm_entry;
1586 rem = current_ir_graph;
1587 obstack_init(&temp_obst);
1589 /* a map for the copied graphs, used to inline recursive calls */
1590 copied_graphs = pmap_create();
1592 /* extend all irgs by a temporary data structure for inlining. */
1593 n_irgs = get_irp_n_irgs();
1594 for (i = 0; i < n_irgs; ++i)
1595 set_irg_link(get_irp_irg(i), alloc_inline_irg_env());
1597 /* Pre-compute information in temporary data structure. */
1598 wenv.ignore_runtime = ignore_runtime;
1599 wenv.ignore_callers = 0;
1600 for (i = 0; i < n_irgs; ++i) {
1601 ir_graph *irg = get_irp_irg(i);
1603 assert(get_irg_phase_state(irg) != phase_building);
1604 free_callee_info(irg);
1606 assure_cf_loop(irg);
1607 wenv.x = get_irg_link(irg);
1608 irg_walk_graph(irg, NULL, collect_calls2, &wenv);
1611 /* -- and now inline. -- */
1613 /* Inline leaves recursively -- we might construct new leaves. */
1617 for (i = 0; i < n_irgs; ++i) {
1619 int phiproj_computed = 0;
1621 current_ir_graph = get_irp_irg(i);
1622 env = get_irg_link(current_ir_graph);
1624 ir_reserve_resources(current_ir_graph, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
1625 list_for_each_entry_safe(call_entry, entry, next, &env->calls, list) {
1627 irg_inline_property prop;
1629 if (env->n_nodes > maxsize)
1633 callee = entry->callee;
1635 prop = get_irg_inline_property(callee);
1636 if (prop == irg_inline_forbidden || get_irg_additional_properties(callee) & mtp_property_weak) {
1637 /* do not inline forbidden / weak graphs */
1641 if (is_leave(callee) && (
1642 is_smaller(callee, leavesize) || prop >= irg_inline_forced)) {
1643 if (!phiproj_computed) {
1644 phiproj_computed = 1;
1645 collect_phiprojs(current_ir_graph);
1647 did_inline = inline_method(call, callee);
1650 inline_irg_env *callee_env = get_irg_link(callee);
1652 /* call was inlined, Phi/Projs for current graph must be recomputed */
1653 phiproj_computed = 0;
1655 /* Do some statistics */
1656 env->got_inline = 1;
1657 --env->n_call_nodes;
1658 env->n_nodes += callee_env->n_nodes;
1659 --callee_env->n_callers;
1661 /* remove this call from the list */
1662 list_del(&entry->list);
1667 ir_free_resources(current_ir_graph, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
1669 } while (did_inline);
1671 /* inline other small functions. */
1672 for (i = 0; i < n_irgs; ++i) {
1674 int phiproj_computed = 0;
1676 current_ir_graph = get_irp_irg(i);
1677 env = get_irg_link(current_ir_graph);
1679 ir_reserve_resources(current_ir_graph, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
1681 /* note that the list of possible calls is updated during the process */
1682 list_for_each_entry_safe(call_entry, entry, next, &env->calls, list) {
1683 irg_inline_property prop;
1688 callee = entry->callee;
1690 prop = get_irg_inline_property(callee);
1691 if (prop == irg_inline_forbidden || get_irg_additional_properties(callee) & mtp_property_weak) {
1692 /* do not inline forbidden / weak graphs */
1696 e = pmap_find(copied_graphs, callee);
1699 * Remap callee if we have a copy.
1700 * FIXME: Should we do this only for recursive Calls ?
1705 if (prop >= irg_inline_forced ||
1706 (is_smaller(callee, size) && env->n_nodes < maxsize) /* small function */) {
1707 if (current_ir_graph == callee) {
1709 * Recursive call: we cannot directly inline because we cannot walk
1710 * the graph and change it. So we have to make a copy of the graph
1714 inline_irg_env *callee_env;
1717 ir_free_resources(current_ir_graph, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
1720 * No copy yet, create one.
1721 * Note that recursive methods are never leaves, so it is sufficient
1722 * to test this condition here.
1724 copy = create_irg_copy(callee);
1726 /* create_irg_copy() destroys the Proj links, recompute them */
1727 phiproj_computed = 0;
1729 ir_reserve_resources(current_ir_graph, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
1731 /* allocate new environment */
1732 callee_env = alloc_inline_irg_env();
1733 set_irg_link(copy, callee_env);
1735 assure_cf_loop(copy);
1736 wenv.x = callee_env;
1737 wenv.ignore_callers = 1;
1738 irg_walk_graph(copy, NULL, collect_calls2, &wenv);
1741 * Enter the entity of the original graph. This is needed
1742 * for inline_method(). However, note that ent->irg still points
1743 * to callee, NOT to copy.
1745 set_irg_entity(copy, get_irg_entity(callee));
1747 pmap_insert(copied_graphs, callee, copy);
1750 /* we have only one caller: the original graph */
1751 callee_env->n_callers = 1;
1752 callee_env->n_callers_orig = 1;
1754 if (! phiproj_computed) {
1755 phiproj_computed = 1;
1756 collect_phiprojs(current_ir_graph);
1758 did_inline = inline_method(call, callee);
1760 inline_irg_env *callee_env = (inline_irg_env *)get_irg_link(callee);
1762 /* call was inlined, Phi/Projs for current graph must be recomputed */
1763 phiproj_computed = 0;
1765 /* callee was inline. Append it's call list. */
1766 env->got_inline = 1;
1767 --env->n_call_nodes;
1768 append_call_list(env, callee_env, entry->loop_depth);
1769 --callee_env->n_callers;
1771 /* after we have inlined callee, all called methods inside callee
1772 are now called once more */
1773 list_for_each_entry(call_entry, centry, &callee_env->calls, list) {
1774 inline_irg_env *penv = get_irg_link(centry->callee);
1778 /* remove this call from the list */
1779 list_del(&entry->list);
1784 ir_free_resources(current_ir_graph, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
1787 for (i = 0; i < n_irgs; ++i) {
1788 irg = get_irp_irg(i);
1789 env = get_irg_link(irg);
1791 if (env->got_inline) {
1792 optimize_graph_df(irg);
1795 if (env->got_inline || (env->n_callers_orig != env->n_callers)) {
1796 DB((dbg, LEVEL_1, "Nodes:%3d ->%3d, calls:%3d ->%3d, callers:%3d ->%3d, -- %s\n",
1797 env->n_nodes_orig, env->n_nodes, env->n_call_nodes_orig, env->n_call_nodes,
1798 env->n_callers_orig, env->n_callers,
1799 get_entity_name(get_irg_entity(irg))));
1803 /* kill the copied graphs: we don't need them anymore */
1804 foreach_pmap(copied_graphs, pm_entry) {
1805 ir_graph *copy = pm_entry->value;
1807 /* reset the entity, otherwise it will be deleted in the next step ... */
1808 set_irg_entity(copy, NULL);
1809 free_ir_graph(copy);
1811 pmap_destroy(copied_graphs);
1813 obstack_free(&temp_obst, NULL);
1814 current_ir_graph = rem;
1817 struct inline_leave_functions_pass_t {
1818 ir_prog_pass_t pass;
1826 * Wrapper to run inline_leave_functions() as a ir_prog pass.
1828 static int inline_leave_functions_wrapper(ir_prog *irp, void *context) {
1829 struct inline_leave_functions_pass_t *pass = context;
1832 inline_leave_functions(
1833 pass->maxsize, pass->leavesize,
1834 pass->size, pass->ignore_runtime);
1838 /* create a pass for inline_leave_functions() */
1839 ir_prog_pass_t *inline_leave_functions_pass(
1840 const char *name, unsigned maxsize, unsigned leavesize,
1841 unsigned size, int ignore_runtime) {
1842 struct inline_leave_functions_pass_t *pass =
1843 XMALLOCZ(struct inline_leave_functions_pass_t);
1845 pass->maxsize = maxsize;
1846 pass->leavesize = leavesize;
1848 pass->ignore_runtime = ignore_runtime;
1850 return def_prog_pass_constructor(
1852 name ? name : "inline_leave_functions",
1853 inline_leave_functions_wrapper);
1857 * Calculate the parameter weights for transmitting the address of a local variable.
1859 static unsigned calc_method_local_weight(ir_node *arg) {
1861 unsigned v, weight = 0;
1863 for (i = get_irn_n_outs(arg) - 1; i >= 0; --i) {
1864 ir_node *succ = get_irn_out(arg, i);
1866 switch (get_irn_opcode(succ)) {
1869 /* Loads and Store can be removed */
1873 /* check if all args are constant */
1874 for (j = get_Sel_n_indexs(succ) - 1; j >= 0; --j) {
1875 ir_node *idx = get_Sel_index(succ, j);
1876 if (! is_Const(idx))
1879 /* Check users on this Sel. Note: if a 0 is returned here, there was
1880 some unsupported node. */
1881 v = calc_method_local_weight(succ);
1884 /* we can kill one Sel with constant indexes, this is cheap */
1888 /* when looking backward we might find Id nodes */
1889 weight += calc_method_local_weight(succ);
1892 /* unoptimized tuple */
1893 for (j = get_Tuple_n_preds(succ) - 1; j >= 0; --j) {
1894 ir_node *pred = get_Tuple_pred(succ, j);
1896 /* look for Proj(j) */
1897 for (k = get_irn_n_outs(succ) - 1; k >= 0; --k) {
1898 ir_node *succ_succ = get_irn_out(succ, k);
1899 if (is_Proj(succ_succ)) {
1900 if (get_Proj_proj(succ_succ) == j) {
1902 weight += calc_method_local_weight(succ_succ);
1905 /* this should NOT happen */
1913 /* any other node: unsupported yet or bad. */
1921 * Calculate the parameter weights for transmitting the address of a local variable.
1923 static void analyze_irg_local_weights(inline_irg_env *env, ir_graph *irg) {
1924 ir_entity *ent = get_irg_entity(irg);
1926 int nparams, i, proj_nr;
1927 ir_node *irg_args, *arg;
1929 mtp = get_entity_type(ent);
1930 nparams = get_method_n_params(mtp);
1932 /* allocate a new array. currently used as 'analysed' flag */
1933 env->local_weights = NEW_ARR_D(unsigned, &temp_obst, nparams);
1935 /* If the method haven't parameters we have nothing to do. */
1939 assure_irg_outs(irg);
1940 irg_args = get_irg_args(irg);
1941 for (i = get_irn_n_outs(irg_args) - 1; i >= 0; --i) {
1942 arg = get_irn_out(irg_args, i);
1943 proj_nr = get_Proj_proj(arg);
1944 env->local_weights[proj_nr] = calc_method_local_weight(arg);
1949 * Calculate the benefice for transmitting an local variable address.
1950 * After inlining, the local variable might be transformed into a
1951 * SSA variable by scalar_replacement().
1953 static unsigned get_method_local_adress_weight(ir_graph *callee, int pos) {
1954 inline_irg_env *env = get_irg_link(callee);
1956 if (env->local_weights != NULL) {
1957 if (pos < ARR_LEN(env->local_weights))
1958 return env->local_weights[pos];
1962 analyze_irg_local_weights(env, callee);
1964 if (pos < ARR_LEN(env->local_weights))
1965 return env->local_weights[pos];
1970 * Calculate a benefice value for inlining the given call.
1972 * @param call the call node we have to inspect
1973 * @param callee the called graph
1975 static int calc_inline_benefice(call_entry *entry, ir_graph *callee)
1977 ir_node *call = entry->call;
1978 ir_entity *ent = get_irg_entity(callee);
1982 int i, n_params, all_const;
1984 irg_inline_property prop;
1986 inline_irg_env *callee_env;
1988 prop = get_irg_inline_property(callee);
1989 if (prop == irg_inline_forbidden) {
1990 DB((dbg, LEVEL_2, "In %+F Call to %+F: inlining forbidden\n",
1992 return entry->benefice = INT_MIN;
1995 if (get_irg_additional_properties(callee) & (mtp_property_noreturn | mtp_property_weak)) {
1996 DB((dbg, LEVEL_2, "In %+F Call to %+F: not inlining noreturn or weak\n",
1998 return entry->benefice = INT_MIN;
2001 /* costs for every passed parameter */
2002 n_params = get_Call_n_params(call);
2003 mtp = get_entity_type(ent);
2004 cc = get_method_calling_convention(mtp);
2005 if (cc & cc_reg_param) {
2006 /* register parameter, smaller costs for register parameters */
2007 int max_regs = cc & ~cc_bits;
2009 if (max_regs < n_params)
2010 weight += max_regs * 2 + (n_params - max_regs) * 5;
2012 weight += n_params * 2;
2014 /* parameters are passed an stack */
2015 weight += 5 * n_params;
2018 /* constant parameters improve the benefice */
2019 frame_ptr = get_irg_frame(current_ir_graph);
2021 for (i = 0; i < n_params; ++i) {
2022 ir_node *param = get_Call_param(call, i);
2024 if (is_Const(param)) {
2025 weight += get_method_param_weight(ent, i);
2028 if (is_SymConst(param))
2029 weight += get_method_param_weight(ent, i);
2030 else if (is_Sel(param) && get_Sel_ptr(param) == frame_ptr) {
2032 * An address of a local variable is transmitted. After
2033 * inlining, scalar_replacement might be able to remove the
2034 * local variable, so honor this.
2036 v = get_method_local_adress_weight(callee, i);
2039 entry->local_adr = 1;
2043 entry->all_const = all_const;
2045 callee_env = get_irg_link(callee);
2046 if (callee_env->n_callers == 1 &&
2047 callee != current_ir_graph &&
2048 get_entity_visibility(ent) == visibility_local) {
2052 /* give a bonus for functions with one block */
2053 if (callee_env->n_blocks == 1)
2054 weight = weight * 3 / 2;
2056 /* and one for small non-recursive functions: we want them to be inlined in mostly every case */
2057 if (callee_env->n_nodes < 30 && !callee_env->recursive)
2060 /* and finally for leaves: they do not increase the register pressure
2061 because of callee safe registers */
2062 if (callee_env->n_call_nodes == 0)
2065 /** it's important to inline inner loops first */
2066 if (entry->loop_depth > 30)
2067 weight += 30 * 1024;
2069 weight += entry->loop_depth * 1024;
2072 * All arguments constant is probably a good sign, give an extra bonus
2077 return entry->benefice = weight;
2080 static ir_graph **irgs;
2081 static int last_irg;
2084 * Callgraph walker, collect all visited graphs.
2086 static void callgraph_walker(ir_graph *irg, void *data) {
2088 irgs[last_irg++] = irg;
2092 * Creates an inline order for all graphs.
2094 * @return the list of graphs.
2096 static ir_graph **create_irg_list(void) {
2097 ir_entity **free_methods;
2099 int n_irgs = get_irp_n_irgs();
2101 cgana(&arr_len, &free_methods);
2102 xfree(free_methods);
2104 compute_callgraph();
2107 irgs = XMALLOCNZ(ir_graph*, n_irgs);
2109 callgraph_walk(NULL, callgraph_walker, NULL);
2110 assert(n_irgs == last_irg);
2116 * Push a call onto the priority list if its benefice is big enough.
2118 * @param pqueue the priority queue of calls
2119 * @param call the call entry
2120 * @param inlien_threshold
2121 * the threshold value
2123 static void maybe_push_call(pqueue_t *pqueue, call_entry *call,
2124 int inline_threshold)
2126 ir_graph *callee = call->callee;
2127 irg_inline_property prop = get_irg_inline_property(callee);
2128 int benefice = calc_inline_benefice(call, callee);
2130 DB((dbg, LEVEL_2, "In %+F Call %+F to %+F has benefice %d\n",
2131 get_irn_irg(call->call), call->call, callee, benefice));
2133 if (prop < irg_inline_forced && benefice < inline_threshold) {
2137 pqueue_put(pqueue, call, benefice);
2141 * Try to inline calls into a graph.
2143 * @param irg the graph into which we inline
2144 * @param maxsize do NOT inline if the size of irg gets
2145 * bigger than this amount
2146 * @param inline_threshold
2147 * threshold value for inline decision
2148 * @param copied_graphs
2149 * map containing copied of recursive graphs
2151 static void inline_into(ir_graph *irg, unsigned maxsize,
2152 int inline_threshold, pmap *copied_graphs)
2154 int phiproj_computed = 0;
2155 inline_irg_env *env = get_irg_link(irg);
2156 call_entry *curr_call;
2160 if (env->n_call_nodes == 0)
2163 if (env->n_nodes > maxsize) {
2164 DB((dbg, LEVEL_2, "%+F: too big (%d)\n", irg, env->n_nodes));
2168 current_ir_graph = irg;
2169 ir_reserve_resources(irg, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
2171 /* put irgs into the pqueue */
2172 pqueue = new_pqueue();
2174 list_for_each_entry(call_entry, curr_call, &env->calls, list) {
2175 assert(is_Call(curr_call->call));
2176 maybe_push_call(pqueue, curr_call, inline_threshold);
2179 /* note that the list of possible calls is updated during the process */
2180 while (!pqueue_empty(pqueue)) {
2182 call_entry *curr_call = pqueue_pop_front(pqueue);
2183 ir_graph *callee = curr_call->callee;
2184 ir_node *call_node = curr_call->call;
2185 inline_irg_env *callee_env = get_irg_link(callee);
2186 irg_inline_property prop = get_irg_inline_property(callee);
2188 const call_entry *centry;
2191 if ((prop < irg_inline_forced) && env->n_nodes + callee_env->n_nodes > maxsize) {
2192 DB((dbg, LEVEL_2, "%+F: too big (%d) + %+F (%d)\n", irg,
2193 env->n_nodes, callee, callee_env->n_nodes));
2197 e = pmap_find(copied_graphs, callee);
2199 int benefice = curr_call->benefice;
2201 * Reduce the weight for recursive function IFF not all arguments are const.
2202 * inlining recursive functions is rarely good.
2204 if (!curr_call->all_const)
2206 if (benefice < inline_threshold)
2210 * Remap callee if we have a copy.
2213 callee_env = get_irg_link(callee);
2216 if (current_ir_graph == callee) {
2218 * Recursive call: we cannot directly inline because we cannot
2219 * walk the graph and change it. So we have to make a copy of
2222 int benefice = curr_call->benefice;
2226 * Reduce the weight for recursive function IFF not all arguments are const.
2227 * inlining recursive functions is rarely good.
2229 if (!curr_call->all_const)
2231 if (benefice < inline_threshold)
2234 ir_free_resources(irg, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
2237 * No copy yet, create one.
2238 * Note that recursive methods are never leaves, so it is
2239 * sufficient to test this condition here.
2241 copy = create_irg_copy(callee);
2243 /* create_irg_copy() destroys the Proj links, recompute them */
2244 phiproj_computed = 0;
2246 ir_reserve_resources(irg, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
2248 /* allocate a new environment */
2249 callee_env = alloc_inline_irg_env();
2250 set_irg_link(copy, callee_env);
2252 assure_cf_loop(copy);
2253 wenv.x = callee_env;
2254 wenv.ignore_callers = 1;
2255 irg_walk_graph(copy, NULL, collect_calls2, &wenv);
2258 * Enter the entity of the original graph. This is needed
2259 * for inline_method(). However, note that ent->irg still points
2260 * to callee, NOT to copy.
2262 set_irg_entity(copy, get_irg_entity(callee));
2264 pmap_insert(copied_graphs, callee, copy);
2267 /* we have only one caller: the original graph */
2268 callee_env->n_callers = 1;
2269 callee_env->n_callers_orig = 1;
2271 if (! phiproj_computed) {
2272 phiproj_computed = 1;
2273 collect_phiprojs(current_ir_graph);
2275 did_inline = inline_method(call_node, callee);
2279 /* call was inlined, Phi/Projs for current graph must be recomputed */
2280 phiproj_computed = 0;
2282 /* remove it from the caller list */
2283 list_del(&curr_call->list);
2285 /* callee was inline. Append it's call list. */
2286 env->got_inline = 1;
2287 if (curr_call->local_adr)
2288 env->local_vars = 1;
2289 --env->n_call_nodes;
2291 /* we just generate a bunch of new calls */
2292 loop_depth = curr_call->loop_depth;
2293 list_for_each_entry(call_entry, centry, &callee_env->calls, list) {
2294 inline_irg_env *penv = get_irg_link(centry->callee);
2296 call_entry *new_entry;
2298 /* after we have inlined callee, all called methods inside
2299 * callee are now called once more */
2302 /* Note that the src list points to Call nodes in the inlined graph,
2303 * but we need Call nodes in our graph. Luckily the inliner leaves
2304 * this information in the link field. */
2305 new_call = get_irn_link(centry->call);
2306 assert(is_Call(new_call));
2308 new_entry = duplicate_call_entry(centry, new_call, loop_depth);
2309 list_add_tail(&new_entry->list, &env->calls);
2310 maybe_push_call(pqueue, new_entry, inline_threshold);
2313 env->n_call_nodes += callee_env->n_call_nodes;
2314 env->n_nodes += callee_env->n_nodes;
2315 --callee_env->n_callers;
2317 ir_free_resources(irg, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
2322 * Heuristic inliner. Calculates a benefice value for every call and inlines
2323 * those calls with a value higher than the threshold.
2325 void inline_functions(unsigned maxsize, int inline_threshold) {
2326 inline_irg_env *env;
2330 pmap *copied_graphs;
2331 pmap_entry *pm_entry;
2334 rem = current_ir_graph;
2335 obstack_init(&temp_obst);
2337 irgs = create_irg_list();
2339 /* a map for the copied graphs, used to inline recursive calls */
2340 copied_graphs = pmap_create();
2342 /* extend all irgs by a temporary data structure for inlining. */
2343 n_irgs = get_irp_n_irgs();
2344 for (i = 0; i < n_irgs; ++i)
2345 set_irg_link(irgs[i], alloc_inline_irg_env());
2347 /* Pre-compute information in temporary data structure. */
2348 wenv.ignore_runtime = 0;
2349 wenv.ignore_callers = 0;
2350 for (i = 0; i < n_irgs; ++i) {
2351 ir_graph *irg = irgs[i];
2353 free_callee_info(irg);
2355 wenv.x = get_irg_link(irg);
2356 assure_cf_loop(irg);
2357 irg_walk_graph(irg, NULL, collect_calls2, &wenv);
2360 /* -- and now inline. -- */
2361 for (i = 0; i < n_irgs; ++i) {
2362 ir_graph *irg = irgs[i];
2364 inline_into(irg, maxsize, inline_threshold, copied_graphs);
2367 for (i = 0; i < n_irgs; ++i) {
2368 ir_graph *irg = irgs[i];
2370 env = get_irg_link(irg);
2371 if (env->got_inline) {
2372 /* this irg got calls inlined: optimize it */
2373 if (get_opt_combo()) {
2374 if (env->local_vars) {
2375 scalar_replacement_opt(irg);
2379 if (env->local_vars) {
2380 if (scalar_replacement_opt(irg)) {
2381 optimize_graph_df(irg);
2387 if (env->got_inline || (env->n_callers_orig != env->n_callers)) {
2388 DB((dbg, LEVEL_1, "Nodes:%3d ->%3d, calls:%3d ->%3d, callers:%3d ->%3d, -- %s\n",
2389 env->n_nodes_orig, env->n_nodes, env->n_call_nodes_orig, env->n_call_nodes,
2390 env->n_callers_orig, env->n_callers,
2391 get_entity_name(get_irg_entity(irg))));
2395 /* kill the copied graphs: we don't need them anymore */
2396 foreach_pmap(copied_graphs, pm_entry) {
2397 ir_graph *copy = pm_entry->value;
2399 /* reset the entity, otherwise it will be deleted in the next step ... */
2400 set_irg_entity(copy, NULL);
2401 free_ir_graph(copy);
2403 pmap_destroy(copied_graphs);
2407 obstack_free(&temp_obst, NULL);
2408 current_ir_graph = rem;
2411 struct inline_functions_pass_t {
2412 ir_prog_pass_t pass;
2414 int inline_threshold;
2418 * Wrapper to run inline_functions() as a ir_prog pass.
2420 static int inline_functions_wrapper(ir_prog *irp, void *context) {
2421 struct inline_functions_pass_t *pass = context;
2424 inline_functions(pass->maxsize, pass->inline_threshold);
2428 /* create a ir_prog pass for inline_functions */
2429 ir_prog_pass_t *inline_functions_pass(
2430 const char *name, unsigned maxsize, int inline_threshold) {
2431 struct inline_functions_pass_t *pass =
2432 XMALLOCZ(struct inline_functions_pass_t);
2434 pass->maxsize = maxsize;
2435 pass->inline_threshold = inline_threshold;
2437 return def_prog_pass_constructor(
2438 &pass->pass, name ? name : "inline_functions",
2439 inline_functions_wrapper);
2442 void firm_init_inline(void) {
2443 FIRM_DBG_REGISTER(dbg, "firm.opt.inline");