X-Git-Url: http://nsz.repo.hu/git/?a=blobdiff_plain;f=ir%2Fana%2Firouts.c;h=c95fd431f064f676b045629e682ba26c4ebfb5c7;hb=7a483ab981d403222150c320242adaad13d60af9;hp=3ebce10d3a640b4590141a6de320f082098e8692;hpb=018347a2cd0639b89c2985759f359fdf02536f79;p=libfirm

diff --git a/ir/ana/irouts.c b/ir/ana/irouts.c
index 3ebce10d3..c95fd431f 100644
--- a/ir/ana/irouts.c
+++ b/ir/ana/irouts.c
@@ -1,31 +1,31 @@
 /*
- * Project:     libFIRM
- * File name:   ir/ana/irouts.c
- * Purpose:     Compute and access out edges.
- * Author:      Goetz Lindenmaier
- * Modified by:
- * Created:     1.2002
- * CVS-ID:      $Id$
- * Copyright:   (c) 2002-2003 Universität Karlsruhe
- * Licence:     This file protected by GPL -  GNU GENERAL PUBLIC LICENSE.
+ * Copyright (C) 1995-2011 University of Karlsruhe.  All right reserved.
+ *
+ * This file is part of libFirm.
+ *
+ * This file may be distributed and/or modified under the terms of the
+ * GNU General Public License version 2 as published by the Free Software
+ * Foundation and appearing in the file LICENSE.GPL included in the
+ * packaging of this file.
+ *
+ * Licensees holding valid libFirm Professional Edition licenses may use
+ * this file in accordance with the libFirm Commercial License.
+ * Agreement provided with the Software.
+ *
+ * This file is provided AS IS with NO WARRANTY OF ANY KIND, INCLUDING THE
+ * WARRANTY OF DESIGN, MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+ * PURPOSE.
  */
 
- /**
-  * @file irouts.c Compute out edges for ir nodes (also called def-use edges).
-  *
-  * Copyright (C) 2002 by Universitaet Karlsruhe
-  * All rights reserved.
-  *
-  * Authors:  Goetz Lindenmaier
-  */
-
-#ifdef HAVE_CONFIG_H
+/**
+ * @file
+ * @brief    Compute and access out edges (also called def-use edges).
+ * @author   Goetz Lindenmaier, Michael Beck
+ * @date     1.2002
+ */
 #include "config.h"
-#endif
 
-#ifdef HAVE_STRING_H
 #include <string.h>
-#endif
 
 #include "xmalloc.h"
 #include "irouts.h"
@@ -33,165 +33,244 @@
 #include "irgraph_t.h"
 #include "irprog_t.h"
 #include "irgwalk.h"
-
-#ifdef DEBUG_libfirm
-/* Note:  ir_node.out_valid and ir_graph.n_outs are only present when DEBUG_libfirm is defined */
-/* Accesses to out_valid and n_outs are fenced out to avoid breakage
-   when compiling with neither DEBUG_libfirm or NDEBUG defined */
-#endif /* defined DEBUG_libfirm */
+#include "util.h"
+#include "irprintf.h"
+#include "error.h"
+#include "ircons.h"
 
 /*--------------------------------------------------------------------*/
 /** Accessing the out datastructures                                 **/
 /*--------------------------------------------------------------------*/
 
+#ifdef DEBUG_libfirm
 /** Clear the outs of a node */
-static void reset_outs (ir_node *node, void *unused)
+static void reset_outs(ir_node *node, void *unused)
+{
+	(void) unused;
+	node->out       = NULL;
+	node->out_valid = 0;
+}
+#endif
+
+int get_irn_outs_computed(const ir_node *node)
 {
-  node->out = NULL;
+	return node->out != NULL;
+}
+
+int get_irn_n_outs(const ir_node *node)
+{
+	assert(node && node->kind == k_ir_node);
 #ifdef DEBUG_libfirm
-  node->out_valid = 0;
+	assert(node->out_valid);
 #endif /* defined DEBUG_libfirm */
+	/* we misuse the first for the size info of the out array */
+	return node->out[0].pos;
 }
 
-/* returns the number of successors of the node: */
-int get_irn_n_outs    (ir_node *node) {
-  assert(node && node->kind == k_ir_node);
+ir_node *get_irn_out(const ir_node *def, int pos)
+{
+	assert(pos >= 0 && pos < get_irn_n_outs(def));
 #ifdef DEBUG_libfirm
-  /* assert (node->out_valid); */
+	assert(def->out_valid);
 #endif /* defined DEBUG_libfirm */
-  return (int)(node->out[0]);
+	return def->out[pos+1].use;
 }
 
-/* Access successor n */
-ir_node *get_irn_out      (ir_node *node, int pos) {
-  assert(pos >= 0 && pos < get_irn_n_outs(node));
+ir_node *get_irn_out_ex(const ir_node *def, int pos, int *in_pos)
+{
+	assert(pos >= 0 && pos < get_irn_n_outs(def));
 #ifdef DEBUG_libfirm
-  /* assert (node->out_valid); */
+	assert(def->out_valid);
 #endif /* defined DEBUG_libfirm */
-  return node->out[pos+1];
+	*in_pos = def->out[pos+1].pos;
+	return def->out[pos+1].use;
 }
 
-void set_irn_out      (ir_node *node, int pos, ir_node *out) {
-  assert(node && out);
-  assert(pos >= 0 && pos < get_irn_n_outs(node));
+void set_irn_out(ir_node *def, int pos, ir_node *use, int in_pos)
+{
+	assert(def && use);
+	assert(pos >= 0 && pos < get_irn_n_outs(def));
 #ifdef DEBUG_libfirm
-  node->out_valid = 1;          /* assume that this function is used correctly */
+	assert(def->out_valid);
 #endif /* defined DEBUG_libfirm */
-  node->out[pos+1] = out;
+	def->out[pos+1].use = use;
+	def->out[pos+1].pos = in_pos;
 }
 
-
-int get_Block_n_cfg_outs (ir_node *bl) {
-  int i, n_cfg_outs = 0;
-  assert(bl && (get_irn_op(bl) == op_Block));
+int get_Block_n_cfg_outs(const ir_node *bl)
+{
+	int i, n_cfg_outs = 0;
+	assert(bl && is_Block(bl));
 #ifdef DEBUG_libfirm
-  assert (bl->out_valid);
+	assert(bl->out_valid);
 #endif /* defined DEBUG_libfirm */
-  for (i = 0; i < (int)bl->out[0]; i++)
-    if ((get_irn_mode(bl->out[i+1]) == mode_X) &&
-        (get_irn_op(bl->out[i+1]) != op_End))
-      n_cfg_outs++;
-  return n_cfg_outs;
+	for (i = 1; i <= bl->out[0].pos; ++i) {
+		ir_node *succ = bl->out[i].use;
+		if (get_irn_mode(succ) == mode_X && !is_End(succ) && !is_Bad(succ))
+			n_cfg_outs += succ->out[0].pos;
+	}
+	return n_cfg_outs;
 }
 
+int get_Block_n_cfg_outs_ka(const ir_node *bl)
+{
+	int i, n_cfg_outs = 0;
+	assert(bl && is_Block(bl));
+#ifdef DEBUG_libfirm
+	assert(bl->out_valid);
+#endif /* defined DEBUG_libfirm */
+	for (i = 1; i <= bl->out[0].pos; ++i) {
+		ir_node *succ = bl->out[i].use;
+		if (get_irn_mode(succ) == mode_X) {
+			if (is_Bad(succ))
+				continue;
+			if (is_End(succ)) {
+				/* ignore End if we are in the Endblock */
+				if (get_nodes_block(succ) == bl)
+					continue;
+				else /* count Keep-alive as one */
+					n_cfg_outs += 1;
+			} else
+				n_cfg_outs += succ->out[0].pos;
+		}
+	}
+	return n_cfg_outs;
+}
 
-ir_node *get_Block_cfg_out  (ir_node *bl, int pos) {
-  int i, out_pos = 0;
-  assert(bl && (get_irn_op(bl) == op_Block));
+ir_node *get_Block_cfg_out(const ir_node *bl, int pos)
+{
+	int i;
+	assert(bl && is_Block(bl));
 #ifdef DEBUG_libfirm
-  assert (bl->out_valid);
+	assert(bl->out_valid);
 #endif /* defined DEBUG_libfirm */
-  for (i = 0; i < (int)bl->out[0]; i++)
-    if ((get_irn_mode(bl->out[i+1]) == mode_X)  &&
-        (get_irn_op(bl->out[i+1]) != op_End)) {
-      if (out_pos == pos) {
-        ir_node *cfop = bl->out[i+1];
-        return cfop->out[0+1];
-      } else {
-        out_pos++;
-      }
-    }
-  return NULL;
+	for (i = 1; i <= bl->out[0].pos; ++i) {
+		ir_node *succ = bl->out[i].use;
+		if (get_irn_mode(succ) == mode_X && !is_End(succ) && !is_Bad(succ)) {
+			int n_outs = succ->out[0].pos;
+			if (pos < n_outs)
+				return succ->out[pos + 1].use;
+			else
+				pos -= n_outs;
+		}
+	}
+	return NULL;
 }
 
-static void irg_out_walk_2(ir_node *node,  irg_walk_func *pre,
-            irg_walk_func *post, void *env) {
-  int i;
-  ir_node *succ;
+ir_node *get_Block_cfg_out_ka(const ir_node *bl, int pos)
+{
+	int i, n_outs;
+	assert(bl && is_Block(bl));
+#ifdef DEBUG_libfirm
+	assert (bl->out_valid);
+#endif /* defined DEBUG_libfirm */
+	for (i = 1; i <= bl->out[0].pos; ++i) {
+		ir_node *succ = bl->out[i].use;
+		if (get_irn_mode(succ) == mode_X) {
+			if (is_Bad(succ))
+				continue;
+			if (is_End(succ)) {
+				ir_node *end_bl = get_nodes_block(succ);
+				if (end_bl == bl) {
+					/* ignore End if we are in the Endblock */
+					continue;
+				}
+				if (pos == 0) {
+					/* handle keep-alive here: return the Endblock instead of the End node */
+					return end_bl;
+				} else
+					--pos;
+			} else {
+				n_outs = succ->out[0].pos;
+				if (pos < n_outs)
+					return succ->out[pos + 1].use;
+				else
+					pos -= n_outs;
+			}
+		}
+	}
+	return NULL;
+}
 
-  assert(node);
-  assert(get_irn_visited(node) < get_irg_visited(current_ir_graph));
+static void irg_out_walk_2(ir_node *node, irg_walk_func *pre,
+                           irg_walk_func *post, void *env)
+{
+	int     i, n;
+	ir_node *succ;
 
-  set_irn_visited(node, get_irg_visited(current_ir_graph));
+	assert(node);
+	assert(get_irn_visited(node) < get_irg_visited(current_ir_graph));
 
-  if (pre) pre(node, env);
+	set_irn_visited(node, get_irg_visited(current_ir_graph));
 
-  for (i = 0; i < get_irn_n_outs(node); i++) {
-    succ = get_irn_out(node, i);
-    if (get_irn_visited(succ) < get_irg_visited(current_ir_graph))
-      irg_out_walk_2(succ, pre, post, env);
-  }
+	if (pre) pre(node, env);
 
-  if (post) post(node, env);
+	for (i = 0, n = get_irn_n_outs(node); i < n; ++i) {
+		succ = get_irn_out(node, i);
+		if (get_irn_visited(succ) < get_irg_visited(current_ir_graph))
+			irg_out_walk_2(succ, pre, post, env);
+	}
 
-  return;
+	if (post) post(node, env);
 }
 
-void irg_out_walk(ir_node *node,
-            irg_walk_func *pre, irg_walk_func *post,
-            void *env) {
-  assert(node);
-  if (get_irg_outs_state(current_ir_graph) != outs_none) {
-    inc_irg_visited (current_ir_graph);
-    irg_out_walk_2(node, pre, post, env);
-  }
-  return;
+void irg_out_walk(ir_node *node, irg_walk_func *pre, irg_walk_func *post,
+                  void *env)
+{
+	assert(node);
+	ir_graph *irg = get_irn_irg(node);
+	if (irg_has_properties(irg, IR_GRAPH_PROPERTY_CONSISTENT_OUTS)) {
+		inc_irg_visited (irg);
+		irg_out_walk_2(node, pre, post, env);
+	}
 }
 
-static void irg_out_block_walk2(ir_node *bl,
-            irg_walk_func *pre, irg_walk_func *post,
-            void *env) {
-  int i;
+static void irg_out_block_walk2(ir_node *bl, irg_walk_func *pre,
+                                irg_walk_func *post, void *env)
+{
+	int i, n;
 
-  if(get_Block_block_visited(bl) < get_irg_block_visited(current_ir_graph)) {
-    set_Block_block_visited(bl, get_irg_block_visited(current_ir_graph));
+	if (!Block_block_visited(bl)) {
+		mark_Block_block_visited(bl);
 
-    if(pre)
-      pre(bl, env);
+		if (pre)
+			pre(bl, env);
 
-    for(i = 0; i < get_Block_n_cfg_outs(bl); i++) {
-      /* find the corresponding predecessor block. */
-      ir_node *pred = get_Block_cfg_out(bl, i);
-      /* recursion */
-      irg_out_block_walk2(pred, pre, post, env);
-    }
+		for (i = 0, n =  get_Block_n_cfg_outs(bl); i < n; ++i) {
+			/* find the corresponding predecessor block. */
+			ir_node *pred = get_Block_cfg_out(bl, i);
+			/* recursion */
+			irg_out_block_walk2(pred, pre, post, env);
+		}
 
-    if(post)
-      post(bl, env);
-  }
-  return;
+		if (post)
+			post(bl, env);
+	}
 }
 
-/* Walks only over Block nodes in the graph.  Has it's own visited
-   flag, so that it can be interleaved with the other walker.         */
-void irg_out_block_walk(ir_node *node,
-            irg_walk_func *pre, irg_walk_func *post,
-            void *env) {
-
-  assert((get_irn_op(node) == op_Block) || (get_irn_mode(node) == mode_X));
-
-  inc_irg_block_visited(current_ir_graph);
+void irg_out_block_walk(ir_node *node, irg_walk_func *pre, irg_walk_func *post,
+                        void *env)
+{
 
-  if (get_irn_mode(node) == mode_X) node = node->out[1];
+	assert(is_Block(node) || (get_irn_mode(node) == mode_X));
 
-  irg_out_block_walk2(node, pre, post, env);
+	inc_irg_block_visited(current_ir_graph);
 
-  return;
+	if (get_irn_mode(node) == mode_X) {
+		int i, n;
 
+		for (i = 0, n = get_irn_n_outs(node); i < n; ++i) {
+			ir_node *succ = get_irn_out(node, i);
+			irg_out_block_walk2(succ, pre, post, env);
+		}
+	}
+	else {
+		irg_out_block_walk2(node, pre, post, env);
+	}
 }
 
 /*--------------------------------------------------------------------*/
-/** Building and Removing the out datasturcture                      **/
+/** Building and Removing the out datastructure                      **/
 /**                                                                  **/
 /** The outs of a graph are allocated in a single, large array.      **/
 /** This allows to allocate and deallocate the memory for the outs   **/
@@ -211,300 +290,212 @@ void irg_out_block_walk(ir_node *node,
 
 
 /** Returns the amount of out edges for not yet visited successors. */
-static int count_outs(ir_node *n) {
-  int start, i, res, irn_arity;
-
-  set_irn_visited(n, get_irg_visited(current_ir_graph));
-  n->out = (ir_node **) 1;     /* Space for array size. */
-
-  start = is_Block(n) ? 0 : -1;
-  irn_arity = get_irn_arity(n);
-  res = irn_arity - start + 1;  /* --1 or --0; 1 for array size. */
-
-  for (i = start; i < irn_arity; i++) {
-    /* Optimize Tuples.  They annoy if walking the cfg. */
-    ir_node *succ = skip_Tuple(get_irn_n(n, i));
-    set_irn_n(n, i, succ);
-
-    /* count outs for successors */
-    if (get_irn_visited(succ) < get_irg_visited(current_ir_graph)) {
-      res += count_outs(succ);
-    }
-    /* Count my outs */
-    succ->out = (ir_node **)( (int)succ->out + 1);
-  }
-  return res;
-}
-
-/**
- * Enter memory for the outs to a node.
- *
- * @param n      current node
- * @param free   current free address in the chunk allocated for the outs
- *
- * @return The next free address
- */
-static ir_node **set_out_edges(ir_node *n, ir_node **free) {
-  int n_outs, start, i, irn_arity;
-  ir_node *succ;
-
-  set_irn_visited(n, get_irg_visited(current_ir_graph));
-
-  /* Allocate my array */
-  n_outs = (int) n->out;
-  n->out = free;
-#ifdef DEBUG_libfirm
-  n->out_valid = 1;
-#endif /* defined DEBUG_libfirm */
-  free += n_outs;
-  /* We count the successors again, the space will be sufficient.
-     We use this counter to remember the position for the next back
-     edge. */
-  n->out[0] = (ir_node *)0;
-
-  start = is_Block(n) ? 0 : -1;
-  irn_arity = get_irn_arity(n);
-
-  for (i = start; i < irn_arity; i++) {
-    succ = get_irn_n(n, i);
-    /* Recursion */
-    if (get_irn_visited(succ) < get_irg_visited(current_ir_graph))
-      free = set_out_edges(succ, free);
-    /* Remember our back edge */
-    succ->out[get_irn_n_outs(succ)+1] = n;
-    succ->out[0] = (ir_node *) (get_irn_n_outs(succ) + 1);
-  }
-  return free;
-}
-
-/* We want that the out of ProjX from Start contains the next block at
-   position 1, the Start block at position 2.  This is necessary for
-   the out block walker. */
-static INLINE void fix_start_proj(ir_graph *irg) {
-  ir_node *proj    = NULL;
-  ir_node *startbl = get_irg_start_block(irg);
-  int i;
-
-  if (get_Block_n_cfg_outs(startbl)) {
-    for (i = 0; i < get_irn_n_outs(startbl); i++)
-      if (get_irn_mode(get_irn_out(startbl, i)) == mode_X) {
-        proj = get_irn_out(startbl, i);
-        break;
-      }
-
-    if (get_irn_out(proj, 0) == startbl) {
-      assert(get_irn_n_outs(proj) == 2);
-      set_irn_out(proj, 0, get_irn_out(proj, 1));
-      set_irn_out(proj, 1, startbl);
-    }
-  }
-}
-
-/* compute the outs for a given graph */
-void compute_outs(ir_graph *irg) {
-  ir_graph *rem = current_ir_graph;
-  int n_out_edges = 0;
-  ir_node **end = NULL;         /* Only for debugging */
-
-  current_ir_graph = irg;
-
-  /* Update graph state */
-  assert(get_irg_phase_state(current_ir_graph) != phase_building);
-
-  if (current_ir_graph->outs_state != outs_none)
-    free_outs(current_ir_graph);
-  current_ir_graph->outs_state = outs_consistent;
+static int _count_outs(ir_node *n)
+{
+	int start, i, res, irn_arity;
 
-  /* This first iteration counts the overall number of out edges and the
-     number of out edges for each node. */
-  inc_irg_visited(irg);
-  n_out_edges = count_outs(get_irg_end(irg));
+	mark_irn_visited(n);
+	n->out = (ir_def_use_edge*) INT_TO_PTR(1);     /* Space for array size. */
 
-  /* allocate memory for all out edges. */
-  irg->outs = xcalloc(n_out_edges, sizeof(irg->outs[0]));
-#ifdef DEBUG_libfirm
-  irg->n_outs = n_out_edges;
-#endif /* defined DEBUG_libfirm */
+	start = is_Block(n) ? 0 : -1;
+	irn_arity = get_irn_arity(n);
+	res = irn_arity - start + 1;  /* --1 or --0; 1 for array size. */
 
-  /* The second iteration splits the irg->outs array into smaller arrays
-     for each node and writes the back edges into this array. */
-  inc_irg_visited(irg);
-  end = set_out_edges(get_irg_end(irg), irg->outs);
+	for (i = start; i < irn_arity; ++i) {
+		/* Optimize Tuples.  They annoy if walking the cfg. */
+		ir_node *pred         = get_irn_n(n, i);
+		ir_node *skipped_pred = skip_Tuple(pred);
 
-  /* Check how much memory we have used */
-  assert (end == (irg->outs + n_out_edges));
+		if (skipped_pred != pred) {
+			set_irn_n(n, i, skipped_pred);
+		}
 
-  /* We want that the out of ProjX from Start contains the next block at
-     position 1, the Start block at position 2.  This is necessary for
-     the out block walker. */
-  fix_start_proj(irg);
+		/* count Def-Use edges for predecessors */
+		if (!irn_visited(skipped_pred))
+			res += _count_outs(skipped_pred);
 
-  current_ir_graph = rem;
+		/*count my Def-Use edges */
+		skipped_pred->out = (ir_def_use_edge*) INT_TO_PTR(PTR_TO_INT(skipped_pred->out) + 1);
+	}
+	return res;
 }
 
 
-
-
-/*------------------------------------------------------------*
- *  This computes the outedges for in interprocedural graph.  *
- *  There is one quirk:                                       *
- *  The number of the outedges for each node is saved in      *
- *  the first member of the ir_node** array. Maybe we should  *
- *  change this to make it more portable...                   *
- *------------------------------------------------------------*/
-
-
-/**
- * Inits the number of outedges for each node
- * before counting.
+/** Returns the amount of out edges for not yet visited successors.
+ *  This version handles some special nodes like irg_frame, irg_args etc.
  */
-static void init_count(ir_node * node, void *env) {
-  node->out = (ir_node **) 1; /* 1 for the array size */
+static int count_outs(ir_graph *irg)
+{
+	ir_node *n;
+	int     i, res;
+
+	inc_irg_visited(irg);
+	res = _count_outs(get_irg_end(irg));
+
+	/* Now handle anchored nodes. We need the out count of those
+	   even if they are not visible. */
+	for (i = anchor_last; i >= anchor_first; --i) {
+		n = get_irg_anchor(irg, i);
+		if (!irn_visited_else_mark(n)) {
+			n->out = (ir_def_use_edge*) INT_TO_PTR(1);
+			++res;
+		}
+	}
+	return res;
 }
 
-
 /**
- * Adjusts the out edge count for its predecessors
- * and adds the current arity to the overall count,
- * which is saved in "env"
+ * Enter memory for the outs to a node.
+ *
+ * @param use    current node
+ * @param free   current free address in the chunk allocated for the outs
+ *
+ * @return The next free address
  */
-static void node_arity_count(ir_node * node, void * env)
+static ir_def_use_edge *_set_out_edges(ir_node *use, ir_def_use_edge *free)
 {
-  int *anz = (int *) env, arity, n_outs, i, start;
-  ir_node *succ;
-
-  arity = get_irn_arity(node);
-  start = (is_Block(node)) ? 0 : -1;
-
-  n_outs = 1 + arity + (-start);  // ((is_Block(node)) ? 0 : 1);   // Why + 1??
-  *anz += n_outs;
-
-  for(i = start; i < arity; i++) {
-    succ = get_irn_n(node, i);
-    succ->out = (ir_node **)((int)succ->out + 1);
-  }
-}
-
-
-/*
- * Inits all nodes for setting the outedges
- * Returns the overall count of edges
- */
-int count_ip_outs(void) {
-
-  int res = 0;
+	int    start, i, irn_arity, pos;
+	size_t n_outs;
 
-  cg_walk(init_count, node_arity_count, &res);
+	mark_irn_visited(use);
 
-  return(res);
+	/* Allocate my array */
+	n_outs = PTR_TO_INT(use->out);
+	use->out = free;
+#ifdef DEBUG_libfirm
+	use->out_valid = 1;
+#endif /* defined DEBUG_libfirm */
+	free += n_outs;
+	/* We count the successors again, the space will be sufficient.
+	   We use this counter to remember the position for the next back
+	   edge. */
+	use->out[0].pos = 0;
+
+	start = is_Block(use) ? 0 : -1;
+	irn_arity = get_irn_arity(use);
+
+	for (i = start; i < irn_arity; ++i) {
+		ir_node *def = get_irn_n(use, i);
+
+		/* Recursion */
+		if (!irn_visited(def))
+			free = _set_out_edges(def, free);
+
+		/* Remember this Def-Use edge */
+		pos = def->out[0].pos + 1;
+		def->out[pos].use = use;
+		def->out[pos].pos = i;
+
+		/* increase the number of Def-Use edges so far */
+		def->out[0].pos = pos;
+	}
+	return free;
 }
 
-static int dummy_count = 0, global_count; /* Only for debugging */
-
 /**
- * For each node: Sets the pointer to array
- * in which the outedges are written later.
- * The current array start is transported in env
+ * Enter memory for the outs to a node. Handles special nodes
+ *
+ * @param irg    the graph
+ * @param free   current free address in the chunk allocated for the outs
+ *
+ * @return The next free address
  */
-static void set_array_pointer(ir_node *node, void *env) {
-
-  int n_outs;
-  ir_node ***free = (ir_node ***) env;
-
-  /* Allocate my array */
-  n_outs = (int) node -> out;  /* We wrote the count here in count_ip_outs */
-  dummy_count += n_outs;
-  assert(dummy_count <= global_count && "More outedges than initially counted!");
-  node -> out = *free;
-  *free = &((*free)[n_outs]);
-  /* We count the successors again, the space will be sufficient.
-     We use this counter to remember the position for the next back
-     edge. */
-  node -> out[0] = (ir_node *) 0;
-}
-
+static ir_def_use_edge *set_out_edges(ir_graph *irg, ir_def_use_edge *free)
+{
+	ir_node *n;
+	int     i;
+
+	inc_irg_visited(irg);
+	free = _set_out_edges(get_irg_end(irg), free);
+
+	/* handle anchored nodes */
+	for (i = anchor_last; i >= anchor_first; --i) {
+		n = get_irg_anchor(irg, i);
+		if (!irn_visited_else_mark(n)) {
+			size_t n_outs = PTR_TO_INT(n->out);
+			n->out = free;
+#ifdef DEBUG_libfirm
+			n->out_valid = 1;
+#endif /* defined DEBUG_libfirm */
+			free += n_outs;
+		}
+	}
 
-/**
- * Adds an outedge from the predecessor to the
- * current node.
- */
-static void set_out_pointer(ir_node * node, void * env) {
-  int i, arity = get_irn_arity(node);
-  ir_node *succ;
-  int start = (!is_Block(node)) ? -1 : 0;
-
-  for(i = start; i < arity; i++) {
-    succ = get_irn_n(node, i);
-    succ->out[get_irn_n_outs(succ)+1] = node;
-    succ->out[0] = (ir_node *) (get_irn_n_outs(succ) + 1);
-  }
+	return free;
 }
 
-
-/*
- * Sets the outedges for all nodes.
- */
-void set_ip_outs(void)
+void compute_irg_outs(ir_graph *irg)
 {
-  ir_node **outedge_array = get_irp_ip_outedges();
-  cg_walk(set_array_pointer, set_out_pointer, (void *) &outedge_array);
-}
+	ir_graph        *rem = current_ir_graph;
+	int             n_out_edges = 0;
+	ir_def_use_edge *end = NULL;         /* Only for debugging */
 
+	current_ir_graph = irg;
 
+	/* Update graph state */
+	assert(get_irg_phase_state(current_ir_graph) != phase_building);
 
-/*
- * Counts the outedges, allocates memory to save the
- * outedges and fills this outedge array in interprocedural
- * view!
- */
-void compute_ip_outs(void) {
+	free_irg_outs(current_ir_graph);
+
+	/* This first iteration counts the overall number of out edges and the
+	   number of out edges for each node. */
+	n_out_edges = count_outs(irg);
 
-  int n_out_edges;
-  ir_node **out_edges;
+	/* allocate memory for all out edges. */
+	irg->outs = XMALLOCNZ(ir_def_use_edge, n_out_edges);
+#ifdef DEBUG_libfirm
+	irg->n_outs = n_out_edges;
+#endif /* defined DEBUG_libfirm */
 
-  assert(get_irp_ip_view_state() == ip_view_valid &&
-     "Cannot construct outs for invalid ip view.");
+	/* The second iteration splits the irg->outs array into smaller arrays
+	   for each node and writes the back edges into this array. */
+	end = set_out_edges(irg, irg->outs);
 
-  if (irp->outs_state != outs_none) {
-    free_ip_outs();
-  }
+	/* Check how much memory we have used */
+	assert (end == (irg->outs + n_out_edges));
 
-  global_count = n_out_edges = count_ip_outs();
-  out_edges = xcalloc(n_out_edges, sizeof(out_edges[0]));
-  set_irp_ip_outedges(out_edges);
-  set_ip_outs();
+	add_irg_properties(irg, IR_GRAPH_PROPERTY_CONSISTENT_OUTS);
+	current_ir_graph = rem;
 }
 
-void free_ip_outs(void)
+void assure_irg_outs(ir_graph *irg)
 {
-  ir_node **out_edges = get_irp_ip_outedges();
-  if (out_edges != NULL) {
-    free(out_edges);
-    set_irp_ip_outedges(NULL);
-  }
-  irp->outs_state = outs_none;
+	if (! irg_has_properties(irg, IR_GRAPH_PROPERTY_CONSISTENT_OUTS))
+		compute_irg_outs(irg);
 }
 
+void compute_irp_outs(void)
+{
+	size_t i, n;
+	for (i = 0, n = get_irp_n_irgs(); i < n; ++i)
+		compute_irg_outs(get_irp_irg(i));
+}
 
-void free_outs(ir_graph *irg) {
+void free_irp_outs(void)
+{
+	size_t i, n;
+	for (i = 0, n = get_irp_n_irgs(); i < n; ++i)
+		free_irg_outs(get_irp_irg(i));
+}
 
-  /*   current_ir_graph->outs_state = outs_none; */
-  irg->outs_state = outs_none;
+void free_irg_outs(ir_graph *irg)
+{
+	/*   current_ir_graph->outs_state = outs_none; */
 
-  if (irg->outs) {
+	if (irg->outs) {
 #ifdef DEBUG_libfirm
-    memset(irg->outs, 0, irg->n_outs);
+		memset(irg->outs, 0, irg->n_outs);
 #endif /* defined DEBUG_libfirm */
-    free(irg->outs);
-    irg->outs = NULL;
+		free(irg->outs);
+		irg->outs = NULL;
 #ifdef DEBUG_libfirm
-    irg->n_outs = 0;
+		irg->n_outs = 0;
 #endif /* defined DEBUG_libfirm */
-  }
+	}
 
 #ifdef DEBUG_libfirm
-  /* when debugging, *always* reset all nodes' outs!  irg->outs might
-     have been lying to us */
-  irg_walk_graph (irg, reset_outs, NULL, NULL);
+	/* when debugging, *always* reset all nodes' outs!  irg->outs might
+	   have been lying to us */
+	irg_walk_graph (irg, reset_outs, NULL, NULL);
 #endif /* defined DEBUG_libfirm */
 }