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 Move nodes to a block where they will be executed the least
24 * @author Christian Schaefer, Goetz Lindenmaier, Sebastian Felis,
27 * The idea here is to push nodes as deep into the dominance tree as their
28 * dependencies allow. After pushing them back up out of as many loops as
35 #include "iroptimize.h"
42 static bool is_block_reachable(ir_node *block)
44 return get_Block_dom_depth(block) >= 0;
48 * Find the earliest correct block for node n. --- Place n into the
49 * same Block as its dominance-deepest Input.
51 * move_out_of_loops() expects that place_floats_early() have placed
52 * all "living" nodes into a living block. That's why we must
53 * move nodes in dead block with "live" successors into a valid
55 * We move them just into the same block as its successor (or
56 * in case of a Phi into the effective use block). For Phi successors,
57 * this may still be a dead block, but then there is no real use, as
58 * the control flow will be dead later.
60 static void place_floats_early(ir_node *n, waitq *worklist)
70 /* we must not run into an infinite loop */
71 if (irn_visited_else_mark(n))
74 /* The algorithm relies on the fact that all predecessors of a block are
75 * moved up after a call to place_float_early of the predecessors
76 * (see the loop below).
77 * However we have to break cycles somewhere. Relying on the visited flag
78 * will result in nodes not being moved up despite their place_floats_early
80 * Instead we break cycles at pinned nodes which won't move anyway:
81 * This works because in firm each cycle contains a Phi or Block node
84 if (get_irn_pinned(n) != op_pin_state_floats) {
85 /* we can't move pinned nodes */
86 arity = get_irn_arity(n);
87 for (i = 0; i < arity; ++i) {
88 ir_node *pred = get_irn_n(n, i);
89 pdeq_putr(worklist, pred);
92 pdeq_putr(worklist, get_nodes_block(n));
96 block = get_nodes_block(n);
98 /* first move predecessors up */
99 arity = get_irn_arity(n);
100 place_floats_early(block, worklist);
101 for (i = 0; i < arity; ++i) {
102 ir_node *pred = get_irn_n(n, i);
103 place_floats_early(pred, worklist);
106 /* determine earliest point */
109 for (i = 0; i < arity; ++i) {
110 ir_node *pred = get_irn_n(n, i);
111 ir_node *pred_block = get_nodes_block(pred);
112 int pred_depth = get_Block_dom_depth(pred_block);
113 if (pred_depth > new_depth) {
114 new_depth = pred_depth;
115 new_block = pred_block;
119 /* avoid moving nodes into the start block if we are not in the backend */
120 irg = get_irn_irg(n);
121 start_block = get_irg_start_block(irg);
122 if (new_block == start_block && block != start_block &&
123 get_irg_phase_state(irg) != phase_backend) {
124 assert(get_Block_n_cfg_outs(start_block) == 1);
125 new_block = get_Block_cfg_out(start_block, 0);
128 /* Set the new block */
129 if (new_block != NULL)
130 set_nodes_block(n, new_block);
134 * Floating nodes form subgraphs that begin at nodes as Const, Load,
135 * Start, Call and that end at op_pin_state_pinned nodes as Store, Call.
136 * Place_early places all floating nodes reachable from its argument through
137 * floating nodes and adds all beginnings at op_pin_state_pinned nodes to the
140 * @param worklist a worklist, used for the algorithm, empty on in/output
142 static void place_early(ir_graph *irg, waitq *worklist)
145 inc_irg_visited(irg);
147 /* this inits the worklist */
148 place_floats_early(get_irg_end(irg), worklist);
150 /* Work the content of the worklist. */
151 while (!waitq_empty(worklist)) {
152 ir_node *n = (ir_node*)waitq_get(worklist);
154 place_floats_early(n, worklist);
156 set_irg_pinned(irg, op_pin_state_pinned);
160 * Compute the deepest common dominator tree ancestor of block and dca.
162 * @param dca the deepest common dominator tree ancestor so far,
164 * @param block a block
166 * @return the deepest common dominator tree ancestor of block and dca
168 static ir_node *calc_dom_dca(ir_node *dca, ir_node *block)
170 assert(block != NULL);
172 /* We found a first legal placement. */
176 /* Find a placement that is dominates both, dca and block. */
177 while (get_Block_dom_depth(block) > get_Block_dom_depth(dca))
178 block = get_Block_idom(block);
180 while (get_Block_dom_depth(dca) > get_Block_dom_depth(block)) {
181 dca = get_Block_idom(dca);
184 while (block != dca) {
185 block = get_Block_idom(block); dca = get_Block_idom(dca);
191 * Deepest common dominance ancestor of DCA and CONSUMER of PRODUCER.
192 * I.e., DCA is the block where we might place PRODUCER.
193 * A data flow edge points from producer to consumer.
195 static ir_node *consumer_dom_dca(ir_node *dca, ir_node *consumer,
198 /* Compute the last block into which we can place a node so that it is
200 if (is_Phi(consumer)) {
201 /* our consumer is a Phi-node, the effective use is in all those
202 blocks through which the Phi-node reaches producer */
203 ir_node *phi_block = get_nodes_block(consumer);
204 int arity = get_irn_arity(consumer);
207 for (i = 0; i < arity; i++) {
208 if (get_Phi_pred(consumer, i) == producer) {
209 ir_node *new_block = get_Block_cfgpred_block(phi_block, i);
210 if (is_Bad(new_block))
213 assert(is_block_reachable(new_block));
214 dca = calc_dom_dca(dca, new_block);
218 dca = calc_dom_dca(dca, get_nodes_block(consumer));
223 static inline int get_block_loop_depth(ir_node *block)
225 return get_loop_depth(get_irn_loop(block));
229 * Move n to a block with less loop depth than its current block. The
230 * new block must be dominated by early.
232 * @param n the node that should be moved
233 * @param early the earliest block we can n move to
235 static void move_out_of_loops(ir_node *n, ir_node *early)
237 ir_node *block = get_nodes_block(n);
238 ir_node *best = block;
239 int best_depth = get_block_loop_depth(best);
241 /* Find the region deepest in the dominator tree dominating
242 dca with the least loop nesting depth, but still dominated
243 by our early placement. */
244 while (block != early) {
245 ir_node *idom = get_Block_idom(block);
246 int idom_depth = get_block_loop_depth(idom);
247 if (idom_depth < best_depth) {
249 best_depth = idom_depth;
253 if (best != get_nodes_block(n))
254 set_nodes_block(n, best);
258 * Calculate the deepest common ancestor in the dominator tree of all nodes'
259 * blocks depending on node; our final placement has to dominate DCA.
261 * @param node the definition node
262 * @param dca the deepest common ancestor block so far, initially
265 * @return the deepest common dominator ancestor of all blocks of node's users
267 static ir_node *get_deepest_common_dom_ancestor(ir_node *node, ir_node *dca)
271 for (i = get_irn_n_outs(node) - 1; i >= 0; --i) {
272 ir_node *succ = get_irn_out(node, i);
274 /* keepalive edges are special and don't respect the dominance */
279 /* Proj nodes are in the same block as node, so
280 * the users of Proj are our users. */
281 dca = get_deepest_common_dom_ancestor(succ, dca);
283 assert(is_block_reachable(get_nodes_block(succ)));
284 dca = consumer_dom_dca(dca, succ, node);
291 * Put all the Proj nodes of a node into a given block.
293 * @param node the mode_T node
294 * @param block the block to put the Proj nodes to
296 static void set_projs_block(ir_node *node, ir_node *block)
300 for (i = get_irn_n_outs(node) - 1; i >= 0; --i) {
301 ir_node *succ = get_irn_out(node, i);
303 assert(is_Proj(succ));
305 if (get_irn_mode(succ) == mode_T) {
306 set_projs_block(succ, block);
308 set_nodes_block(succ, block);
313 * Find the latest legal block for N and place N into the
314 * `optimal' Block between the latest and earliest legal block.
315 * The `optimal' block is the dominance-deepest block of those
316 * with the least loop-nesting-depth. This places N out of as many
317 * loops as possible and then makes it as control dependent as
320 static void place_floats_late(ir_node *n, pdeq *worklist)
327 if (irn_visited_else_mark(n))
330 n_outs = get_irn_n_outs(n);
331 /* break cycles at pinned nodes (see place place_floats_early) as to why */
332 if (get_irn_pinned(n) != op_pin_state_floats) {
333 for (i = 0; i < n_outs; ++i) {
334 ir_node *succ = get_irn_out(n, i);
335 pdeq_putr(worklist, succ);
340 /* place our users */
341 for (i = 0; i < n_outs; ++i) {
342 ir_node *succ = get_irn_out(n, i);
343 place_floats_late(succ, worklist);
346 /* no point in moving Projs around, they are moved with their predecessor */
349 /* some nodes should simply stay in the startblock */
350 if (is_irn_start_block_placed(n)) {
351 assert(get_nodes_block(n) == get_irg_start_block(get_irn_irg(n)));
355 block = get_nodes_block(n);
356 assert(is_block_reachable(block));
358 /* deepest common ancestor in the dominator tree of all nodes'
359 blocks depending on us; our final placement has to dominate
361 dca = get_deepest_common_dom_ancestor(n, NULL);
363 set_nodes_block(n, dca);
364 move_out_of_loops(n, block);
365 if (get_irn_mode(n) == mode_T) {
366 set_projs_block(n, get_nodes_block(n));
372 * Place floating nodes on the given worklist as late as possible using
373 * the dominance tree.
375 * @param worklist the worklist containing the nodes to place
377 static void place_late(ir_graph *irg, waitq *worklist)
380 inc_irg_visited(irg);
382 /* This fills the worklist initially. */
383 place_floats_late(get_irg_start_block(irg), worklist);
385 /* And now empty the worklist again... */
386 while (!waitq_empty(worklist)) {
387 ir_node *n = (ir_node*)waitq_get(worklist);
389 place_floats_late(n, worklist);
393 /* Code Placement. */
394 void place_code(ir_graph *irg)
398 /* Handle graph state */
399 assert(get_irg_phase_state(irg) != phase_building);
400 assure_irg_properties(irg,
401 IR_GRAPH_PROPERTY_NO_CRITICAL_EDGES |
402 IR_GRAPH_PROPERTY_NO_UNREACHABLE_CODE |
403 IR_GRAPH_PROPERTY_CONSISTENT_OUTS |
404 IR_GRAPH_PROPERTY_CONSISTENT_DOMINANCE |
405 IR_GRAPH_PROPERTY_CONSISTENT_LOOPINFO);
407 /* Place all floating nodes as early as possible. This guarantees
408 a legal code placement. */
409 worklist = new_waitq();
410 place_early(irg, worklist);
412 /* While GCSE might place nodes in unreachable blocks,
413 * these are now placed in reachable blocks. */
415 /* Note: place_early changes only blocks, no data edges. So, the
416 * data out edges are still valid, no need to recalculate them here. */
418 /* Now move the nodes down in the dominator tree. This reduces the
419 unnecessary executions of the node. */
420 place_late(irg, worklist);
423 confirm_irg_properties(irg, IR_GRAPH_PROPERTIES_NONE);
427 * Wrapper for place_code() inside the place_code pass.
429 static void place_code_wrapper(ir_graph *irg)
431 set_opt_global_cse(1);
432 optimize_graph_df(irg);
434 set_opt_global_cse(0);
437 ir_graph_pass_t *place_code_pass(const char *name)
439 return def_graph_pass(name ? name : "place", place_code_wrapper);