1 /*
   2  * Copyright (c) 1999, 2019, Oracle and/or its affiliates. All rights reserved.
   3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   4  *
   5  * This code is free software; you can redistribute it and/or modify it
   6  * under the terms of the GNU General Public License version 2 only, as
   7  * published by the Free Software Foundation.
   8  *
   9  * This code is distributed in the hope that it will be useful, but WITHOUT
  10  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  11  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  12  * version 2 for more details (a copy is included in the LICENSE file that
  13  * accompanied this code).
  14  *
  15  * You should have received a copy of the GNU General Public License version
  16  * 2 along with this work; if not, write to the Free Software Foundation,
  17  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
  18  *
  19  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
  20  * or visit www.oracle.com if you need additional information or have any
  21  * questions.
  22  *
  23  */
  24 
  25 #include "precompiled.hpp"
  26 #include "gc/shared/barrierSet.hpp"
  27 #include "gc/shared/c2/barrierSetC2.hpp"
  28 #include "memory/allocation.inline.hpp"
  29 #include "memory/resourceArea.hpp"
  30 #include "opto/addnode.hpp"
  31 #include "opto/callnode.hpp"
  32 #include "opto/castnode.hpp"
  33 #include "opto/connode.hpp"
  34 #include "opto/castnode.hpp"
  35 #include "opto/divnode.hpp"
  36 #include "opto/loopnode.hpp"
  37 #include "opto/matcher.hpp"
  38 #include "opto/mulnode.hpp"
  39 #include "opto/movenode.hpp"
  40 #include "opto/opaquenode.hpp"
  41 #include "opto/rootnode.hpp"
  42 #include "opto/subnode.hpp"
  43 #include "utilities/macros.hpp"
  44 #if INCLUDE_ZGC
  45 #include "gc/z/c2/zBarrierSetC2.hpp"
  46 #endif
  47 
  48 //=============================================================================
  49 //------------------------------split_thru_phi---------------------------------
  50 // Split Node 'n' through merge point if there is enough win.
  51 Node *PhaseIdealLoop::split_thru_phi( Node *n, Node *region, int policy ) {
  52   if (n->Opcode() == Op_ConvI2L && n->bottom_type() != TypeLong::LONG) {
  53     // ConvI2L may have type information on it which is unsafe to push up
  54     // so disable this for now
  55     return NULL;
  56   }
  57 
  58   // Splitting range check CastIIs through a loop induction Phi can
  59   // cause new Phis to be created that are left unrelated to the loop
  60   // induction Phi and prevent optimizations (vectorization)
  61   if (n->Opcode() == Op_CastII && n->as_CastII()->has_range_check() &&
  62       region->is_CountedLoop() && n->in(1) == region->as_CountedLoop()->phi()) {
  63     return NULL;
  64   }
  65 
  66   int wins = 0;
  67   assert(!n->is_CFG(), "");
  68   assert(region->is_Region(), "");
  69 
  70   const Type* type = n->bottom_type();
  71   const TypeOopPtr *t_oop = _igvn.type(n)->isa_oopptr();
  72   Node *phi;
  73   if (t_oop != NULL && t_oop->is_known_instance_field()) {
  74     int iid    = t_oop->instance_id();
  75     int index  = C->get_alias_index(t_oop);
  76     int offset = t_oop->offset();
  77     phi = new PhiNode(region, type, NULL, iid, index, offset);
  78   } else {
  79     phi = PhiNode::make_blank(region, n);
  80   }
  81   uint old_unique = C->unique();
  82   for (uint i = 1; i < region->req(); i++) {
  83     Node *x;
  84     Node* the_clone = NULL;
  85     if (region->in(i) == C->top()) {
  86       x = C->top();             // Dead path?  Use a dead data op
  87     } else {
  88       x = n->clone();           // Else clone up the data op
  89       the_clone = x;            // Remember for possible deletion.
  90       // Alter data node to use pre-phi inputs
  91       if (n->in(0) == region)
  92         x->set_req( 0, region->in(i) );
  93       for (uint j = 1; j < n->req(); j++) {
  94         Node *in = n->in(j);
  95         if (in->is_Phi() && in->in(0) == region)
  96           x->set_req( j, in->in(i) ); // Use pre-Phi input for the clone
  97       }
  98     }
  99     // Check for a 'win' on some paths
 100     const Type *t = x->Value(&_igvn);
 101 
 102     bool singleton = t->singleton();
 103 
 104     // A TOP singleton indicates that there are no possible values incoming
 105     // along a particular edge. In most cases, this is OK, and the Phi will
 106     // be eliminated later in an Ideal call. However, we can't allow this to
 107     // happen if the singleton occurs on loop entry, as the elimination of
 108     // the PhiNode may cause the resulting node to migrate back to a previous
 109     // loop iteration.
 110     if (singleton && t == Type::TOP) {
 111       // Is_Loop() == false does not confirm the absence of a loop (e.g., an
 112       // irreducible loop may not be indicated by an affirmative is_Loop());
 113       // therefore, the only top we can split thru a phi is on a backedge of
 114       // a loop.
 115       singleton &= region->is_Loop() && (i != LoopNode::EntryControl);
 116     }
 117 
 118     if (singleton) {
 119       wins++;
 120       x = ((PhaseGVN&)_igvn).makecon(t);
 121     } else {
 122       // We now call Identity to try to simplify the cloned node.
 123       // Note that some Identity methods call phase->type(this).
 124       // Make sure that the type array is big enough for
 125       // our new node, even though we may throw the node away.
 126       // (Note: This tweaking with igvn only works because x is a new node.)
 127       _igvn.set_type(x, t);
 128       // If x is a TypeNode, capture any more-precise type permanently into Node
 129       // otherwise it will be not updated during igvn->transform since
 130       // igvn->type(x) is set to x->Value() already.
 131       x->raise_bottom_type(t);
 132       Node *y = _igvn.apply_identity(x);
 133       if (y != x) {
 134         wins++;
 135         x = y;
 136       } else {
 137         y = _igvn.hash_find(x);
 138         if (y) {
 139           wins++;
 140           x = y;
 141         } else {
 142           // Else x is a new node we are keeping
 143           // We do not need register_new_node_with_optimizer
 144           // because set_type has already been called.
 145           _igvn._worklist.push(x);
 146         }
 147       }
 148     }
 149     if (x != the_clone && the_clone != NULL)
 150       _igvn.remove_dead_node(the_clone);
 151     phi->set_req( i, x );
 152   }
 153   // Too few wins?
 154   if (wins <= policy) {
 155     _igvn.remove_dead_node(phi);
 156     return NULL;
 157   }
 158 
 159   // Record Phi
 160   register_new_node( phi, region );
 161 
 162   for (uint i2 = 1; i2 < phi->req(); i2++) {
 163     Node *x = phi->in(i2);
 164     // If we commoned up the cloned 'x' with another existing Node,
 165     // the existing Node picks up a new use.  We need to make the
 166     // existing Node occur higher up so it dominates its uses.
 167     Node *old_ctrl;
 168     IdealLoopTree *old_loop;
 169 
 170     if (x->is_Con()) {
 171       // Constant's control is always root.
 172       set_ctrl(x, C->root());
 173       continue;
 174     }
 175     // The occasional new node
 176     if (x->_idx >= old_unique) {     // Found a new, unplaced node?
 177       old_ctrl = NULL;
 178       old_loop = NULL;               // Not in any prior loop
 179     } else {
 180       old_ctrl = get_ctrl(x);
 181       old_loop = get_loop(old_ctrl); // Get prior loop
 182     }
 183     // New late point must dominate new use
 184     Node *new_ctrl = dom_lca(old_ctrl, region->in(i2));
 185     if (new_ctrl == old_ctrl) // Nothing is changed
 186       continue;
 187 
 188     IdealLoopTree *new_loop = get_loop(new_ctrl);
 189 
 190     // Don't move x into a loop if its uses are
 191     // outside of loop. Otherwise x will be cloned
 192     // for each use outside of this loop.
 193     IdealLoopTree *use_loop = get_loop(region);
 194     if (!new_loop->is_member(use_loop) &&
 195         (old_loop == NULL || !new_loop->is_member(old_loop))) {
 196       // Take early control, later control will be recalculated
 197       // during next iteration of loop optimizations.
 198       new_ctrl = get_early_ctrl(x);
 199       new_loop = get_loop(new_ctrl);
 200     }
 201     // Set new location
 202     set_ctrl(x, new_ctrl);
 203     // If changing loop bodies, see if we need to collect into new body
 204     if (old_loop != new_loop) {
 205       if (old_loop && !old_loop->_child)
 206         old_loop->_body.yank(x);
 207       if (!new_loop->_child)
 208         new_loop->_body.push(x);  // Collect body info
 209     }
 210   }
 211 
 212   return phi;
 213 }
 214 
 215 //------------------------------dominated_by------------------------------------
 216 // Replace the dominated test with an obvious true or false.  Place it on the
 217 // IGVN worklist for later cleanup.  Move control-dependent data Nodes on the
 218 // live path up to the dominating control.
 219 void PhaseIdealLoop::dominated_by( Node *prevdom, Node *iff, bool flip, bool exclude_loop_predicate ) {
 220   if (VerifyLoopOptimizations && PrintOpto) { tty->print_cr("dominating test"); }
 221 
 222   // prevdom is the dominating projection of the dominating test.
 223   assert( iff->is_If(), "" );
 224   assert(iff->Opcode() == Op_If || iff->Opcode() == Op_CountedLoopEnd || iff->Opcode() == Op_RangeCheck, "Check this code when new subtype is added");
 225   int pop = prevdom->Opcode();
 226   assert( pop == Op_IfFalse || pop == Op_IfTrue, "" );
 227   if (flip) {
 228     if (pop == Op_IfTrue)
 229       pop = Op_IfFalse;
 230     else
 231       pop = Op_IfTrue;
 232   }
 233   // 'con' is set to true or false to kill the dominated test.
 234   Node *con = _igvn.makecon(pop == Op_IfTrue ? TypeInt::ONE : TypeInt::ZERO);
 235   set_ctrl(con, C->root()); // Constant gets a new use
 236   // Hack the dominated test
 237   _igvn.replace_input_of(iff, 1, con);
 238 
 239   // If I dont have a reachable TRUE and FALSE path following the IfNode then
 240   // I can assume this path reaches an infinite loop.  In this case it's not
 241   // important to optimize the data Nodes - either the whole compilation will
 242   // be tossed or this path (and all data Nodes) will go dead.
 243   if (iff->outcnt() != 2) return;
 244 
 245   // Make control-dependent data Nodes on the live path (path that will remain
 246   // once the dominated IF is removed) become control-dependent on the
 247   // dominating projection.
 248   Node* dp = iff->as_If()->proj_out_or_null(pop == Op_IfTrue);
 249 
 250   // Loop predicates may have depending checks which should not
 251   // be skipped. For example, range check predicate has two checks
 252   // for lower and upper bounds.
 253   if (dp == NULL)
 254     return;
 255 
 256   ProjNode* dp_proj  = dp->as_Proj();
 257   ProjNode* unc_proj = iff->as_If()->proj_out(1 - dp_proj->_con)->as_Proj();
 258   if (exclude_loop_predicate &&
 259       (unc_proj->is_uncommon_trap_proj(Deoptimization::Reason_predicate) != NULL ||
 260        unc_proj->is_uncommon_trap_proj(Deoptimization::Reason_profile_predicate) != NULL ||
 261        unc_proj->is_uncommon_trap_proj(Deoptimization::Reason_range_check) != NULL)) {
 262     // If this is a range check (IfNode::is_range_check), do not
 263     // reorder because Compile::allow_range_check_smearing might have
 264     // changed the check.
 265     return; // Let IGVN transformation change control dependence.
 266   }
 267 
 268   IdealLoopTree *old_loop = get_loop(dp);
 269 
 270   for (DUIterator_Fast imax, i = dp->fast_outs(imax); i < imax; i++) {
 271     Node* cd = dp->fast_out(i); // Control-dependent node
 272     if (cd->depends_only_on_test()) {
 273       assert(cd->in(0) == dp, "");
 274       _igvn.replace_input_of(cd, 0, prevdom);
 275       set_early_ctrl(cd);
 276       IdealLoopTree *new_loop = get_loop(get_ctrl(cd));
 277       if (old_loop != new_loop) {
 278         if (!old_loop->_child) old_loop->_body.yank(cd);
 279         if (!new_loop->_child) new_loop->_body.push(cd);
 280       }
 281       --i;
 282       --imax;
 283     }
 284   }
 285 }
 286 
 287 //------------------------------has_local_phi_input----------------------------
 288 // Return TRUE if 'n' has Phi inputs from its local block and no other
 289 // block-local inputs (all non-local-phi inputs come from earlier blocks)
 290 Node *PhaseIdealLoop::has_local_phi_input( Node *n ) {
 291   Node *n_ctrl = get_ctrl(n);
 292   // See if some inputs come from a Phi in this block, or from before
 293   // this block.
 294   uint i;
 295   for( i = 1; i < n->req(); i++ ) {
 296     Node *phi = n->in(i);
 297     if( phi->is_Phi() && phi->in(0) == n_ctrl )
 298       break;
 299   }
 300   if( i >= n->req() )
 301     return NULL;                // No Phi inputs; nowhere to clone thru
 302 
 303   // Check for inputs created between 'n' and the Phi input.  These
 304   // must split as well; they have already been given the chance
 305   // (courtesy of a post-order visit) and since they did not we must
 306   // recover the 'cost' of splitting them by being very profitable
 307   // when splitting 'n'.  Since this is unlikely we simply give up.
 308   for( i = 1; i < n->req(); i++ ) {
 309     Node *m = n->in(i);
 310     if( get_ctrl(m) == n_ctrl && !m->is_Phi() ) {
 311       // We allow the special case of AddP's with no local inputs.
 312       // This allows us to split-up address expressions.
 313       if (m->is_AddP() &&
 314           get_ctrl(m->in(2)) != n_ctrl &&
 315           get_ctrl(m->in(3)) != n_ctrl) {
 316         // Move the AddP up to dominating point
 317         Node* c = find_non_split_ctrl(idom(n_ctrl));
 318         if (c->is_OuterStripMinedLoop()) {
 319           c->as_Loop()->verify_strip_mined(1);
 320           c = c->in(LoopNode::EntryControl);
 321         }
 322         set_ctrl_and_loop(m, c);
 323         continue;
 324       }
 325       return NULL;
 326     }
 327     assert(n->is_Phi() || m->is_Phi() || is_dominator(get_ctrl(m), n_ctrl), "m has strange control");
 328   }
 329 
 330   return n_ctrl;
 331 }
 332 
 333 //------------------------------remix_address_expressions----------------------
 334 // Rework addressing expressions to get the most loop-invariant stuff
 335 // moved out.  We'd like to do all associative operators, but it's especially
 336 // important (common) to do address expressions.
 337 Node *PhaseIdealLoop::remix_address_expressions( Node *n ) {
 338   if (!has_ctrl(n))  return NULL;
 339   Node *n_ctrl = get_ctrl(n);
 340   IdealLoopTree *n_loop = get_loop(n_ctrl);
 341 
 342   // See if 'n' mixes loop-varying and loop-invariant inputs and
 343   // itself is loop-varying.
 344 
 345   // Only interested in binary ops (and AddP)
 346   if( n->req() < 3 || n->req() > 4 ) return NULL;
 347 
 348   Node *n1_ctrl = get_ctrl(n->in(                    1));
 349   Node *n2_ctrl = get_ctrl(n->in(                    2));
 350   Node *n3_ctrl = get_ctrl(n->in(n->req() == 3 ? 2 : 3));
 351   IdealLoopTree *n1_loop = get_loop( n1_ctrl );
 352   IdealLoopTree *n2_loop = get_loop( n2_ctrl );
 353   IdealLoopTree *n3_loop = get_loop( n3_ctrl );
 354 
 355   // Does one of my inputs spin in a tighter loop than self?
 356   if( (n_loop->is_member( n1_loop ) && n_loop != n1_loop) ||
 357       (n_loop->is_member( n2_loop ) && n_loop != n2_loop) ||
 358       (n_loop->is_member( n3_loop ) && n_loop != n3_loop) )
 359     return NULL;                // Leave well enough alone
 360 
 361   // Is at least one of my inputs loop-invariant?
 362   if( n1_loop == n_loop &&
 363       n2_loop == n_loop &&
 364       n3_loop == n_loop )
 365     return NULL;                // No loop-invariant inputs
 366 
 367 
 368   int n_op = n->Opcode();
 369 
 370   // Replace expressions like ((V+I) << 2) with (V<<2 + I<<2).
 371   if( n_op == Op_LShiftI ) {
 372     // Scale is loop invariant
 373     Node *scale = n->in(2);
 374     Node *scale_ctrl = get_ctrl(scale);
 375     IdealLoopTree *scale_loop = get_loop(scale_ctrl );
 376     if( n_loop == scale_loop || !scale_loop->is_member( n_loop ) )
 377       return NULL;
 378     const TypeInt *scale_t = scale->bottom_type()->isa_int();
 379     if( scale_t && scale_t->is_con() && scale_t->get_con() >= 16 )
 380       return NULL;              // Dont bother with byte/short masking
 381     // Add must vary with loop (else shift would be loop-invariant)
 382     Node *add = n->in(1);
 383     Node *add_ctrl = get_ctrl(add);
 384     IdealLoopTree *add_loop = get_loop(add_ctrl);
 385     //assert( n_loop == add_loop, "" );
 386     if( n_loop != add_loop ) return NULL;  // happens w/ evil ZKM loops
 387 
 388     // Convert I-V into I+ (0-V); same for V-I
 389     if( add->Opcode() == Op_SubI &&
 390         _igvn.type( add->in(1) ) != TypeInt::ZERO ) {
 391       Node *zero = _igvn.intcon(0);
 392       set_ctrl(zero, C->root());
 393       Node *neg = new SubINode( _igvn.intcon(0), add->in(2) );
 394       register_new_node( neg, get_ctrl(add->in(2) ) );
 395       add = new AddINode( add->in(1), neg );
 396       register_new_node( add, add_ctrl );
 397     }
 398     if( add->Opcode() != Op_AddI ) return NULL;
 399     // See if one add input is loop invariant
 400     Node *add_var = add->in(1);
 401     Node *add_var_ctrl = get_ctrl(add_var);
 402     IdealLoopTree *add_var_loop = get_loop(add_var_ctrl );
 403     Node *add_invar = add->in(2);
 404     Node *add_invar_ctrl = get_ctrl(add_invar);
 405     IdealLoopTree *add_invar_loop = get_loop(add_invar_ctrl );
 406     if( add_var_loop == n_loop ) {
 407     } else if( add_invar_loop == n_loop ) {
 408       // Swap to find the invariant part
 409       add_invar = add_var;
 410       add_invar_ctrl = add_var_ctrl;
 411       add_invar_loop = add_var_loop;
 412       add_var = add->in(2);
 413       Node *add_var_ctrl = get_ctrl(add_var);
 414       IdealLoopTree *add_var_loop = get_loop(add_var_ctrl );
 415     } else                      // Else neither input is loop invariant
 416       return NULL;
 417     if( n_loop == add_invar_loop || !add_invar_loop->is_member( n_loop ) )
 418       return NULL;              // No invariant part of the add?
 419 
 420     // Yes!  Reshape address expression!
 421     Node *inv_scale = new LShiftINode( add_invar, scale );
 422     Node *inv_scale_ctrl =
 423       dom_depth(add_invar_ctrl) > dom_depth(scale_ctrl) ?
 424       add_invar_ctrl : scale_ctrl;
 425     register_new_node( inv_scale, inv_scale_ctrl );
 426     Node *var_scale = new LShiftINode( add_var, scale );
 427     register_new_node( var_scale, n_ctrl );
 428     Node *var_add = new AddINode( var_scale, inv_scale );
 429     register_new_node( var_add, n_ctrl );
 430     _igvn.replace_node( n, var_add );
 431     return var_add;
 432   }
 433 
 434   // Replace (I+V) with (V+I)
 435   if( n_op == Op_AddI ||
 436       n_op == Op_AddL ||
 437       n_op == Op_AddF ||
 438       n_op == Op_AddD ||
 439       n_op == Op_MulI ||
 440       n_op == Op_MulL ||
 441       n_op == Op_MulF ||
 442       n_op == Op_MulD ) {
 443     if( n2_loop == n_loop ) {
 444       assert( n1_loop != n_loop, "" );
 445       n->swap_edges(1, 2);
 446     }
 447   }
 448 
 449   // Replace ((I1 +p V) +p I2) with ((I1 +p I2) +p V),
 450   // but not if I2 is a constant.
 451   if( n_op == Op_AddP ) {
 452     if( n2_loop == n_loop && n3_loop != n_loop ) {
 453       if( n->in(2)->Opcode() == Op_AddP && !n->in(3)->is_Con() ) {
 454         Node *n22_ctrl = get_ctrl(n->in(2)->in(2));
 455         Node *n23_ctrl = get_ctrl(n->in(2)->in(3));
 456         IdealLoopTree *n22loop = get_loop( n22_ctrl );
 457         IdealLoopTree *n23_loop = get_loop( n23_ctrl );
 458         if( n22loop != n_loop && n22loop->is_member(n_loop) &&
 459             n23_loop == n_loop ) {
 460           Node *add1 = new AddPNode( n->in(1), n->in(2)->in(2), n->in(3) );
 461           // Stuff new AddP in the loop preheader
 462           register_new_node( add1, n_loop->_head->in(LoopNode::EntryControl) );
 463           Node *add2 = new AddPNode( n->in(1), add1, n->in(2)->in(3) );
 464           register_new_node( add2, n_ctrl );
 465           _igvn.replace_node( n, add2 );
 466           return add2;
 467         }
 468       }
 469     }
 470 
 471     // Replace (I1 +p (I2 + V)) with ((I1 +p I2) +p V)
 472     if (n2_loop != n_loop && n3_loop == n_loop) {
 473       if (n->in(3)->Opcode() == Op_AddX) {
 474         Node *V = n->in(3)->in(1);
 475         Node *I = n->in(3)->in(2);
 476         if (is_member(n_loop,get_ctrl(V))) {
 477         } else {
 478           Node *tmp = V; V = I; I = tmp;
 479         }
 480         if (!is_member(n_loop,get_ctrl(I))) {
 481           Node *add1 = new AddPNode(n->in(1), n->in(2), I);
 482           // Stuff new AddP in the loop preheader
 483           register_new_node(add1, n_loop->_head->in(LoopNode::EntryControl));
 484           Node *add2 = new AddPNode(n->in(1), add1, V);
 485           register_new_node(add2, n_ctrl);
 486           _igvn.replace_node(n, add2);
 487           return add2;
 488         }
 489       }
 490     }
 491   }
 492 
 493   return NULL;
 494 }
 495 
 496 // Optimize ((in1[2*i] * in2[2*i]) + (in1[2*i+1] * in2[2*i+1]))
 497 Node *PhaseIdealLoop::convert_add_to_muladd(Node* n) {
 498   assert(n->Opcode() == Op_AddI, "sanity");
 499   Node * nn = NULL;
 500   Node * in1 = n->in(1);
 501   Node * in2 = n->in(2);
 502   if (in1->Opcode() == Op_MulI && in2->Opcode() == Op_MulI) {
 503     IdealLoopTree* loop_n = get_loop(get_ctrl(n));
 504     if (loop_n->_head->as_Loop()->is_valid_counted_loop() &&
 505         Matcher::match_rule_supported(Op_MulAddS2I) &&
 506         Matcher::match_rule_supported(Op_MulAddVS2VI)) {
 507       Node* mul_in1 = in1->in(1);
 508       Node* mul_in2 = in1->in(2);
 509       Node* mul_in3 = in2->in(1);
 510       Node* mul_in4 = in2->in(2);
 511       if (mul_in1->Opcode() == Op_LoadS &&
 512           mul_in2->Opcode() == Op_LoadS &&
 513           mul_in3->Opcode() == Op_LoadS &&
 514           mul_in4->Opcode() == Op_LoadS) {
 515         IdealLoopTree* loop1 = get_loop(get_ctrl(mul_in1));
 516         IdealLoopTree* loop2 = get_loop(get_ctrl(mul_in2));
 517         IdealLoopTree* loop3 = get_loop(get_ctrl(mul_in3));
 518         IdealLoopTree* loop4 = get_loop(get_ctrl(mul_in4));
 519         IdealLoopTree* loop5 = get_loop(get_ctrl(in1));
 520         IdealLoopTree* loop6 = get_loop(get_ctrl(in2));
 521         // All nodes should be in the same counted loop.
 522         if (loop_n == loop1 && loop_n == loop2 && loop_n == loop3 &&
 523             loop_n == loop4 && loop_n == loop5 && loop_n == loop6) {
 524           Node* adr1 = mul_in1->in(MemNode::Address);
 525           Node* adr2 = mul_in2->in(MemNode::Address);
 526           Node* adr3 = mul_in3->in(MemNode::Address);
 527           Node* adr4 = mul_in4->in(MemNode::Address);
 528           if (adr1->is_AddP() && adr2->is_AddP() && adr3->is_AddP() && adr4->is_AddP()) {
 529             if ((adr1->in(AddPNode::Base) == adr3->in(AddPNode::Base)) &&
 530                 (adr2->in(AddPNode::Base) == adr4->in(AddPNode::Base))) {
 531               nn = new MulAddS2INode(mul_in1, mul_in2, mul_in3, mul_in4);
 532               register_new_node(nn, get_ctrl(n));
 533               _igvn.replace_node(n, nn);
 534               return nn;
 535             } else if ((adr1->in(AddPNode::Base) == adr4->in(AddPNode::Base)) &&
 536                        (adr2->in(AddPNode::Base) == adr3->in(AddPNode::Base))) {
 537               nn = new MulAddS2INode(mul_in1, mul_in2, mul_in4, mul_in3);
 538               register_new_node(nn, get_ctrl(n));
 539               _igvn.replace_node(n, nn);
 540               return nn;
 541             }
 542           }
 543         }
 544       }
 545     }
 546   }
 547   return nn;
 548 }
 549 
 550 //------------------------------conditional_move-------------------------------
 551 // Attempt to replace a Phi with a conditional move.  We have some pretty
 552 // strict profitability requirements.  All Phis at the merge point must
 553 // be converted, so we can remove the control flow.  We need to limit the
 554 // number of c-moves to a small handful.  All code that was in the side-arms
 555 // of the CFG diamond is now speculatively executed.  This code has to be
 556 // "cheap enough".  We are pretty much limited to CFG diamonds that merge
 557 // 1 or 2 items with a total of 1 or 2 ops executed speculatively.
 558 Node *PhaseIdealLoop::conditional_move( Node *region ) {
 559 
 560   assert(region->is_Region(), "sanity check");
 561   if (region->req() != 3) return NULL;
 562 
 563   // Check for CFG diamond
 564   Node *lp = region->in(1);
 565   Node *rp = region->in(2);
 566   if (!lp || !rp) return NULL;
 567   Node *lp_c = lp->in(0);
 568   if (lp_c == NULL || lp_c != rp->in(0) || !lp_c->is_If()) return NULL;
 569   IfNode *iff = lp_c->as_If();
 570 
 571   // Check for ops pinned in an arm of the diamond.
 572   // Can't remove the control flow in this case
 573   if (lp->outcnt() > 1) return NULL;
 574   if (rp->outcnt() > 1) return NULL;
 575 
 576   IdealLoopTree* r_loop = get_loop(region);
 577   assert(r_loop == get_loop(iff), "sanity");
 578   // Always convert to CMOVE if all results are used only outside this loop.
 579   bool used_inside_loop = (r_loop == _ltree_root);
 580 
 581   // Check profitability
 582   int cost = 0;
 583   int phis = 0;
 584   for (DUIterator_Fast imax, i = region->fast_outs(imax); i < imax; i++) {
 585     Node *out = region->fast_out(i);
 586     if (!out->is_Phi()) continue; // Ignore other control edges, etc
 587     phis++;
 588     PhiNode* phi = out->as_Phi();
 589     BasicType bt = phi->type()->basic_type();
 590     switch (bt) {
 591     case T_DOUBLE:
 592     case T_FLOAT:
 593       if (C->use_cmove()) {
 594         continue; //TODO: maybe we want to add some cost
 595       }
 596       cost += Matcher::float_cmove_cost(); // Could be very expensive
 597       break;
 598     case T_LONG: {
 599       cost += Matcher::long_cmove_cost(); // May encodes as 2 CMOV's
 600     }
 601     case T_INT:                 // These all CMOV fine
 602     case T_ADDRESS: {           // (RawPtr)
 603       cost++;
 604       break;
 605     }
 606     case T_NARROWOOP: // Fall through
 607     case T_OBJECT: {            // Base oops are OK, but not derived oops
 608       const TypeOopPtr *tp = phi->type()->make_ptr()->isa_oopptr();
 609       // Derived pointers are Bad (tm): what's the Base (for GC purposes) of a
 610       // CMOVE'd derived pointer?  It's a CMOVE'd derived base.  Thus
 611       // CMOVE'ing a derived pointer requires we also CMOVE the base.  If we
 612       // have a Phi for the base here that we convert to a CMOVE all is well
 613       // and good.  But if the base is dead, we'll not make a CMOVE.  Later
 614       // the allocator will have to produce a base by creating a CMOVE of the
 615       // relevant bases.  This puts the allocator in the business of
 616       // manufacturing expensive instructions, generally a bad plan.
 617       // Just Say No to Conditionally-Moved Derived Pointers.
 618       if (tp && tp->offset() != 0)
 619         return NULL;
 620       cost++;
 621       break;
 622     }
 623     default:
 624       return NULL;              // In particular, can't do memory or I/O
 625     }
 626     // Add in cost any speculative ops
 627     for (uint j = 1; j < region->req(); j++) {
 628       Node *proj = region->in(j);
 629       Node *inp = phi->in(j);
 630       if (get_ctrl(inp) == proj) { // Found local op
 631         cost++;
 632         // Check for a chain of dependent ops; these will all become
 633         // speculative in a CMOV.
 634         for (uint k = 1; k < inp->req(); k++)
 635           if (get_ctrl(inp->in(k)) == proj)
 636             cost += ConditionalMoveLimit; // Too much speculative goo
 637       }
 638     }
 639     // See if the Phi is used by a Cmp or Narrow oop Decode/Encode.
 640     // This will likely Split-If, a higher-payoff operation.
 641     for (DUIterator_Fast kmax, k = phi->fast_outs(kmax); k < kmax; k++) {
 642       Node* use = phi->fast_out(k);
 643       if (use->is_Cmp() || use->is_DecodeNarrowPtr() || use->is_EncodeNarrowPtr())
 644         cost += ConditionalMoveLimit;
 645       // Is there a use inside the loop?
 646       // Note: check only basic types since CMoveP is pinned.
 647       if (!used_inside_loop && is_java_primitive(bt)) {
 648         IdealLoopTree* u_loop = get_loop(has_ctrl(use) ? get_ctrl(use) : use);
 649         if (r_loop == u_loop || r_loop->is_member(u_loop)) {
 650           used_inside_loop = true;
 651         }
 652       }
 653     }
 654   }//for
 655   Node* bol = iff->in(1);
 656   assert(bol->Opcode() == Op_Bool, "");
 657   int cmp_op = bol->in(1)->Opcode();
 658   // It is expensive to generate flags from a float compare.
 659   // Avoid duplicated float compare.
 660   if (phis > 1 && (cmp_op == Op_CmpF || cmp_op == Op_CmpD)) return NULL;
 661 
 662   float infrequent_prob = PROB_UNLIKELY_MAG(3);
 663   // Ignore cost and blocks frequency if CMOVE can be moved outside the loop.
 664   if (used_inside_loop) {
 665     if (cost >= ConditionalMoveLimit) return NULL; // Too much goo
 666 
 667     // BlockLayoutByFrequency optimization moves infrequent branch
 668     // from hot path. No point in CMOV'ing in such case (110 is used
 669     // instead of 100 to take into account not exactness of float value).
 670     if (BlockLayoutByFrequency) {
 671       infrequent_prob = MAX2(infrequent_prob, (float)BlockLayoutMinDiamondPercentage/110.0f);
 672     }
 673   }
 674   // Check for highly predictable branch.  No point in CMOV'ing if
 675   // we are going to predict accurately all the time.
 676   if (C->use_cmove() && (cmp_op == Op_CmpF || cmp_op == Op_CmpD)) {
 677     //keep going
 678   } else if (iff->_prob < infrequent_prob ||
 679       iff->_prob > (1.0f - infrequent_prob))
 680     return NULL;
 681 
 682   // --------------
 683   // Now replace all Phis with CMOV's
 684   Node *cmov_ctrl = iff->in(0);
 685   uint flip = (lp->Opcode() == Op_IfTrue);
 686   Node_List wq;
 687   while (1) {
 688     PhiNode* phi = NULL;
 689     for (DUIterator_Fast imax, i = region->fast_outs(imax); i < imax; i++) {
 690       Node *out = region->fast_out(i);
 691       if (out->is_Phi()) {
 692         phi = out->as_Phi();
 693         break;
 694       }
 695     }
 696     if (phi == NULL)  break;
 697     if (PrintOpto && VerifyLoopOptimizations) { tty->print_cr("CMOV"); }
 698     // Move speculative ops
 699     wq.push(phi);
 700     while (wq.size() > 0) {
 701       Node *n = wq.pop();
 702       for (uint j = 1; j < n->req(); j++) {
 703         Node* m = n->in(j);
 704         if (m != NULL && !is_dominator(get_ctrl(m), cmov_ctrl)) {
 705 #ifndef PRODUCT
 706           if (PrintOpto && VerifyLoopOptimizations) {
 707             tty->print("  speculate: ");
 708             m->dump();
 709           }
 710 #endif
 711           set_ctrl(m, cmov_ctrl);
 712           wq.push(m);
 713         }
 714       }
 715     }
 716     Node *cmov = CMoveNode::make(cmov_ctrl, iff->in(1), phi->in(1+flip), phi->in(2-flip), _igvn.type(phi));
 717     register_new_node( cmov, cmov_ctrl );
 718     _igvn.replace_node( phi, cmov );
 719 #ifndef PRODUCT
 720     if (TraceLoopOpts) {
 721       tty->print("CMOV  ");
 722       r_loop->dump_head();
 723       if (Verbose) {
 724         bol->in(1)->dump(1);
 725         cmov->dump(1);
 726       }
 727     }
 728     if (VerifyLoopOptimizations) verify();
 729 #endif
 730   }
 731 
 732   // The useless CFG diamond will fold up later; see the optimization in
 733   // RegionNode::Ideal.
 734   _igvn._worklist.push(region);
 735 
 736   return iff->in(1);
 737 }
 738 
 739 static void enqueue_cfg_uses(Node* m, Unique_Node_List& wq) {
 740   for (DUIterator_Fast imax, i = m->fast_outs(imax); i < imax; i++) {
 741     Node* u = m->fast_out(i);
 742     if (u->is_CFG()) {
 743       if (u->Opcode() == Op_NeverBranch) {
 744         u = ((NeverBranchNode*)u)->proj_out(0);
 745         enqueue_cfg_uses(u, wq);
 746       } else {
 747         wq.push(u);
 748       }
 749     }
 750   }
 751 }
 752 
 753 // Try moving a store out of a loop, right before the loop
 754 Node* PhaseIdealLoop::try_move_store_before_loop(Node* n, Node *n_ctrl) {
 755   // Store has to be first in the loop body
 756   IdealLoopTree *n_loop = get_loop(n_ctrl);
 757   if (n->is_Store() && n_loop != _ltree_root &&
 758       n_loop->is_loop() && n_loop->_head->is_Loop() &&
 759       n->in(0) != NULL) {
 760     Node* address = n->in(MemNode::Address);
 761     Node* value = n->in(MemNode::ValueIn);
 762     Node* mem = n->in(MemNode::Memory);
 763     IdealLoopTree* address_loop = get_loop(get_ctrl(address));
 764     IdealLoopTree* value_loop = get_loop(get_ctrl(value));
 765 
 766     // - address and value must be loop invariant
 767     // - memory must be a memory Phi for the loop
 768     // - Store must be the only store on this memory slice in the
 769     // loop: if there's another store following this one then value
 770     // written at iteration i by the second store could be overwritten
 771     // at iteration i+n by the first store: it's not safe to move the
 772     // first store out of the loop
 773     // - nothing must observe the memory Phi: it guarantees no read
 774     // before the store, we are also guaranteed the store post
 775     // dominates the loop head (ignoring a possible early
 776     // exit). Otherwise there would be extra Phi involved between the
 777     // loop's Phi and the store.
 778     // - there must be no early exit from the loop before the Store
 779     // (such an exit most of the time would be an extra use of the
 780     // memory Phi but sometimes is a bottom memory Phi that takes the
 781     // store as input).
 782 
 783     if (!n_loop->is_member(address_loop) &&
 784         !n_loop->is_member(value_loop) &&
 785         mem->is_Phi() && mem->in(0) == n_loop->_head &&
 786         mem->outcnt() == 1 &&
 787         mem->in(LoopNode::LoopBackControl) == n) {
 788 
 789       assert(n_loop->_tail != NULL, "need a tail");
 790       assert(is_dominator(n_ctrl, n_loop->_tail), "store control must not be in a branch in the loop");
 791 
 792       // Verify that there's no early exit of the loop before the store.
 793       bool ctrl_ok = false;
 794       {
 795         // Follow control from loop head until n, we exit the loop or
 796         // we reach the tail
 797         ResourceMark rm;
 798         Unique_Node_List wq;
 799         wq.push(n_loop->_head);
 800 
 801         for (uint next = 0; next < wq.size(); ++next) {
 802           Node *m = wq.at(next);
 803           if (m == n->in(0)) {
 804             ctrl_ok = true;
 805             continue;
 806           }
 807           assert(!has_ctrl(m), "should be CFG");
 808           if (!n_loop->is_member(get_loop(m)) || m == n_loop->_tail) {
 809             ctrl_ok = false;
 810             break;
 811           }
 812           enqueue_cfg_uses(m, wq);
 813           if (wq.size() > 10) {
 814             ctrl_ok = false;
 815             break;
 816           }
 817         }
 818       }
 819       if (ctrl_ok) {
 820         // move the Store
 821         _igvn.replace_input_of(mem, LoopNode::LoopBackControl, mem);
 822         _igvn.replace_input_of(n, 0, n_loop->_head->as_Loop()->skip_strip_mined()->in(LoopNode::EntryControl));
 823         _igvn.replace_input_of(n, MemNode::Memory, mem->in(LoopNode::EntryControl));
 824         // Disconnect the phi now. An empty phi can confuse other
 825         // optimizations in this pass of loop opts.
 826         _igvn.replace_node(mem, mem->in(LoopNode::EntryControl));
 827         n_loop->_body.yank(mem);
 828 
 829         set_ctrl_and_loop(n, n->in(0));
 830 
 831         return n;
 832       }
 833     }
 834   }
 835   return NULL;
 836 }
 837 
 838 // Try moving a store out of a loop, right after the loop
 839 void PhaseIdealLoop::try_move_store_after_loop(Node* n) {
 840   if (n->is_Store() && n->in(0) != NULL) {
 841     Node *n_ctrl = get_ctrl(n);
 842     IdealLoopTree *n_loop = get_loop(n_ctrl);
 843     // Store must be in a loop
 844     if (n_loop != _ltree_root && !n_loop->_irreducible) {
 845       Node* address = n->in(MemNode::Address);
 846       Node* value = n->in(MemNode::ValueIn);
 847       IdealLoopTree* address_loop = get_loop(get_ctrl(address));
 848       // address must be loop invariant
 849       if (!n_loop->is_member(address_loop)) {
 850         // Store must be last on this memory slice in the loop and
 851         // nothing in the loop must observe it
 852         Node* phi = NULL;
 853         for (DUIterator_Fast imax, i = n->fast_outs(imax); i < imax; i++) {
 854           Node* u = n->fast_out(i);
 855           if (has_ctrl(u)) { // control use?
 856             IdealLoopTree *u_loop = get_loop(get_ctrl(u));
 857             if (!n_loop->is_member(u_loop)) {
 858               continue;
 859             }
 860             if (u->is_Phi() && u->in(0) == n_loop->_head) {
 861               assert(_igvn.type(u) == Type::MEMORY, "bad phi");
 862               // multiple phis on the same slice are possible
 863               if (phi != NULL) {
 864                 return;
 865               }
 866               phi = u;
 867               continue;
 868             }
 869           }
 870           return;
 871         }
 872         if (phi != NULL) {
 873           // Nothing in the loop before the store (next iteration)
 874           // must observe the stored value
 875           bool mem_ok = true;
 876           {
 877             ResourceMark rm;
 878             Unique_Node_List wq;
 879             wq.push(phi);
 880             for (uint next = 0; next < wq.size() && mem_ok; ++next) {
 881               Node *m = wq.at(next);
 882               for (DUIterator_Fast imax, i = m->fast_outs(imax); i < imax && mem_ok; i++) {
 883                 Node* u = m->fast_out(i);
 884                 if (u->is_Store() || u->is_Phi()) {
 885                   if (u != n) {
 886                     wq.push(u);
 887                     mem_ok = (wq.size() <= 10);
 888                   }
 889                 } else {
 890                   mem_ok = false;
 891                   break;
 892                 }
 893               }
 894             }
 895           }
 896           if (mem_ok) {
 897             // Move the store out of the loop if the LCA of all
 898             // users (except for the phi) is outside the loop.
 899             Node* hook = new Node(1);
 900             _igvn.rehash_node_delayed(phi);
 901             int count = phi->replace_edge(n, hook);
 902             assert(count > 0, "inconsistent phi");
 903 
 904             // Compute latest point this store can go
 905             Node* lca = get_late_ctrl(n, get_ctrl(n));
 906             if (n_loop->is_member(get_loop(lca))) {
 907               // LCA is in the loop - bail out
 908               _igvn.replace_node(hook, n);
 909               return;
 910             }
 911 #ifdef ASSERT
 912             if (n_loop->_head->is_Loop() && n_loop->_head->as_Loop()->is_strip_mined()) {
 913               assert(n_loop->_head->Opcode() == Op_CountedLoop, "outer loop is a strip mined");
 914               n_loop->_head->as_Loop()->verify_strip_mined(1);
 915               Node* outer = n_loop->_head->as_CountedLoop()->outer_loop();
 916               IdealLoopTree* outer_loop = get_loop(outer);
 917               assert(n_loop->_parent == outer_loop, "broken loop tree");
 918               assert(get_loop(lca) == outer_loop, "safepoint in outer loop consume all memory state");
 919             }
 920 #endif
 921 
 922             // Move store out of the loop
 923             _igvn.replace_node(hook, n->in(MemNode::Memory));
 924             _igvn.replace_input_of(n, 0, lca);
 925             set_ctrl_and_loop(n, lca);
 926 
 927             // Disconnect the phi now. An empty phi can confuse other
 928             // optimizations in this pass of loop opts..
 929             if (phi->in(LoopNode::LoopBackControl) == phi) {
 930               _igvn.replace_node(phi, phi->in(LoopNode::EntryControl));
 931               n_loop->_body.yank(phi);
 932             }
 933           }
 934         }
 935       }
 936     }
 937   }
 938 }
 939 
 940 //------------------------------split_if_with_blocks_pre-----------------------
 941 // Do the real work in a non-recursive function.  Data nodes want to be
 942 // cloned in the pre-order so they can feed each other nicely.
 943 Node *PhaseIdealLoop::split_if_with_blocks_pre( Node *n ) {
 944   BarrierSetC2* bs = BarrierSet::barrier_set()->barrier_set_c2();
 945   Node* bs_res = bs->split_if_pre(this, n);
 946   if (bs_res != NULL) {
 947     return bs_res;
 948   }
 949   // Cloning these guys is unlikely to win
 950   int n_op = n->Opcode();
 951   if( n_op == Op_MergeMem ) return n;
 952   if( n->is_Proj() ) return n;
 953   // Do not clone-up CmpFXXX variations, as these are always
 954   // followed by a CmpI
 955   if( n->is_Cmp() ) return n;
 956   // Attempt to use a conditional move instead of a phi/branch
 957   if( ConditionalMoveLimit > 0 && n_op == Op_Region ) {
 958     Node *cmov = conditional_move( n );
 959     if( cmov ) return cmov;
 960   }
 961   if( n->is_CFG() || n->is_LoadStore() )
 962     return n;
 963   if( n_op == Op_Opaque1 ||     // Opaque nodes cannot be mod'd
 964       n_op == Op_Opaque2 ) {
 965     if( !C->major_progress() )   // If chance of no more loop opts...
 966       _igvn._worklist.push(n);  // maybe we'll remove them
 967     return n;
 968   }
 969 
 970   if( n->is_Con() ) return n;   // No cloning for Con nodes
 971 
 972   Node *n_ctrl = get_ctrl(n);
 973   if( !n_ctrl ) return n;       // Dead node
 974 
 975   Node* res = try_move_store_before_loop(n, n_ctrl);
 976   if (res != NULL) {
 977     return n;
 978   }
 979 
 980   // Attempt to remix address expressions for loop invariants
 981   Node *m = remix_address_expressions( n );
 982   if( m ) return m;
 983 
 984   if (n_op == Op_AddI) {
 985     Node *nn = convert_add_to_muladd( n );
 986     if ( nn ) return nn;
 987   }
 988 
 989   if (n->is_ConstraintCast()) {
 990     Node* dom_cast = n->as_ConstraintCast()->dominating_cast(&_igvn, this);
 991     // ConstraintCastNode::dominating_cast() uses node control input to determine domination.
 992     // Node control inputs don't necessarily agree with loop control info (due to
 993     // transformations happened in between), thus additional dominance check is needed
 994     // to keep loop info valid.
 995     if (dom_cast != NULL && is_dominator(get_ctrl(dom_cast), get_ctrl(n))) {
 996       _igvn.replace_node(n, dom_cast);
 997       return dom_cast;
 998     }
 999   }
1000 
1001   // Determine if the Node has inputs from some local Phi.
1002   // Returns the block to clone thru.
1003   Node *n_blk = has_local_phi_input( n );
1004   if( !n_blk ) return n;
1005 
1006   // Do not clone the trip counter through on a CountedLoop
1007   // (messes up the canonical shape).
1008   if( n_blk->is_CountedLoop() && n->Opcode() == Op_AddI ) return n;
1009 
1010   // Check for having no control input; not pinned.  Allow
1011   // dominating control.
1012   if (n->in(0)) {
1013     Node *dom = idom(n_blk);
1014     if (dom_lca(n->in(0), dom) != n->in(0)) {
1015       return n;
1016     }
1017   }
1018   // Policy: when is it profitable.  You must get more wins than
1019   // policy before it is considered profitable.  Policy is usually 0,
1020   // so 1 win is considered profitable.  Big merges will require big
1021   // cloning, so get a larger policy.
1022   int policy = n_blk->req() >> 2;
1023 
1024   // If the loop is a candidate for range check elimination,
1025   // delay splitting through it's phi until a later loop optimization
1026   if (n_blk->is_CountedLoop()) {
1027     IdealLoopTree *lp = get_loop(n_blk);
1028     if (lp && lp->_rce_candidate) {
1029       return n;
1030     }
1031   }
1032 
1033   if (must_throttle_split_if()) return n;
1034 
1035   // Split 'n' through the merge point if it is profitable
1036   Node *phi = split_thru_phi( n, n_blk, policy );
1037   if (!phi) return n;
1038 
1039   // Found a Phi to split thru!
1040   // Replace 'n' with the new phi
1041   _igvn.replace_node( n, phi );
1042   // Moved a load around the loop, 'en-registering' something.
1043   if (n_blk->is_Loop() && n->is_Load() &&
1044       !phi->in(LoopNode::LoopBackControl)->is_Load())
1045     C->set_major_progress();
1046 
1047   return phi;
1048 }
1049 
1050 static bool merge_point_too_heavy(Compile* C, Node* region) {
1051   // Bail out if the region and its phis have too many users.
1052   int weight = 0;
1053   for (DUIterator_Fast imax, i = region->fast_outs(imax); i < imax; i++) {
1054     weight += region->fast_out(i)->outcnt();
1055   }
1056   int nodes_left = C->max_node_limit() - C->live_nodes();
1057   if (weight * 8 > nodes_left) {
1058     if (PrintOpto) {
1059       tty->print_cr("*** Split-if bails out:  %d nodes, region weight %d", C->unique(), weight);
1060     }
1061     return true;
1062   } else {
1063     return false;
1064   }
1065 }
1066 
1067 static bool merge_point_safe(Node* region) {
1068   // 4799512: Stop split_if_with_blocks from splitting a block with a ConvI2LNode
1069   // having a PhiNode input. This sidesteps the dangerous case where the split
1070   // ConvI2LNode may become TOP if the input Value() does not
1071   // overlap the ConvI2L range, leaving a node which may not dominate its
1072   // uses.
1073   // A better fix for this problem can be found in the BugTraq entry, but
1074   // expediency for Mantis demands this hack.
1075   // 6855164: If the merge point has a FastLockNode with a PhiNode input, we stop
1076   // split_if_with_blocks from splitting a block because we could not move around
1077   // the FastLockNode.
1078   for (DUIterator_Fast imax, i = region->fast_outs(imax); i < imax; i++) {
1079     Node* n = region->fast_out(i);
1080     if (n->is_Phi()) {
1081       for (DUIterator_Fast jmax, j = n->fast_outs(jmax); j < jmax; j++) {
1082         Node* m = n->fast_out(j);
1083         if (m->is_FastLock())
1084           return false;
1085 #ifdef _LP64
1086         if (m->Opcode() == Op_ConvI2L)
1087           return false;
1088         if (m->is_CastII() && m->isa_CastII()->has_range_check()) {
1089           return false;
1090         }
1091 #endif
1092       }
1093     }
1094   }
1095   return true;
1096 }
1097 
1098 
1099 //------------------------------place_near_use---------------------------------
1100 // Place some computation next to use but not inside inner loops.
1101 // For inner loop uses move it to the preheader area.
1102 Node *PhaseIdealLoop::place_near_use(Node *useblock) const {
1103   IdealLoopTree *u_loop = get_loop( useblock );
1104   if (u_loop->_irreducible) {
1105     return useblock;
1106   }
1107   if (u_loop->_child) {
1108     if (useblock == u_loop->_head && u_loop->_head->is_OuterStripMinedLoop()) {
1109       return u_loop->_head->in(LoopNode::EntryControl);
1110     }
1111     return useblock;
1112   }
1113   return u_loop->_head->as_Loop()->skip_strip_mined()->in(LoopNode::EntryControl);
1114 }
1115 
1116 
1117 bool PhaseIdealLoop::identical_backtoback_ifs(Node *n) {
1118   if (!n->is_If() || n->is_CountedLoopEnd()) {
1119     return false;
1120   }
1121   if (!n->in(0)->is_Region()) {
1122     return false;
1123   }
1124   Node* region = n->in(0);
1125   Node* dom = idom(region);
1126   if (!dom->is_If() || dom->in(1) != n->in(1)) {
1127     return false;
1128   }
1129   IfNode* dom_if = dom->as_If();
1130   Node* proj_true = dom_if->proj_out(1);
1131   Node* proj_false = dom_if->proj_out(0);
1132 
1133   for (uint i = 1; i < region->req(); i++) {
1134     if (is_dominator(proj_true, region->in(i))) {
1135       continue;
1136     }
1137     if (is_dominator(proj_false, region->in(i))) {
1138       continue;
1139     }
1140     return false;
1141   }
1142 
1143   return true;
1144 }
1145 
1146 
1147 bool PhaseIdealLoop::can_split_if(Node* n_ctrl) {
1148   if (must_throttle_split_if()) {
1149     return false;
1150   }
1151 
1152   // Do not do 'split-if' if irreducible loops are present.
1153   if (_has_irreducible_loops) {
1154     return false;
1155   }
1156 
1157   if (merge_point_too_heavy(C, n_ctrl)) {
1158     return false;
1159   }
1160 
1161   // Do not do 'split-if' if some paths are dead.  First do dead code
1162   // elimination and then see if its still profitable.
1163   for (uint i = 1; i < n_ctrl->req(); i++) {
1164     if (n_ctrl->in(i) == C->top()) {
1165       return false;
1166     }
1167   }
1168 
1169   // If trying to do a 'Split-If' at the loop head, it is only
1170   // profitable if the cmp folds up on BOTH paths.  Otherwise we
1171   // risk peeling a loop forever.
1172 
1173   // CNC - Disabled for now.  Requires careful handling of loop
1174   // body selection for the cloned code.  Also, make sure we check
1175   // for any input path not being in the same loop as n_ctrl.  For
1176   // irreducible loops we cannot check for 'n_ctrl->is_Loop()'
1177   // because the alternative loop entry points won't be converted
1178   // into LoopNodes.
1179   IdealLoopTree *n_loop = get_loop(n_ctrl);
1180   for (uint j = 1; j < n_ctrl->req(); j++) {
1181     if (get_loop(n_ctrl->in(j)) != n_loop) {
1182       return false;
1183     }
1184   }
1185 
1186   // Check for safety of the merge point.
1187   if (!merge_point_safe(n_ctrl)) {
1188     return false;
1189   }
1190 
1191   return true;
1192 }
1193 
1194 //------------------------------split_if_with_blocks_post----------------------
1195 // Do the real work in a non-recursive function.  CFG hackery wants to be
1196 // in the post-order, so it can dirty the I-DOM info and not use the dirtied
1197 // info.
1198 void PhaseIdealLoop::split_if_with_blocks_post(Node *n) {
1199 
1200   // Cloning Cmp through Phi's involves the split-if transform.
1201   // FastLock is not used by an If
1202   if (n->is_Cmp() && !n->is_FastLock()) {
1203     Node *n_ctrl = get_ctrl(n);
1204     // Determine if the Node has inputs from some local Phi.
1205     // Returns the block to clone thru.
1206     Node *n_blk = has_local_phi_input(n);
1207     if (n_blk != n_ctrl) {
1208       return;
1209     }
1210 
1211     if (!can_split_if(n_ctrl)) {
1212       return;
1213     }
1214 
1215     if (n->outcnt() != 1) {
1216       return; // Multiple bool's from 1 compare?
1217     }
1218     Node *bol = n->unique_out();
1219     assert(bol->is_Bool(), "expect a bool here");
1220     if (bol->outcnt() != 1) {
1221       return;// Multiple branches from 1 compare?
1222     }
1223     Node *iff = bol->unique_out();
1224 
1225     // Check some safety conditions
1226     if (iff->is_If()) {        // Classic split-if?
1227       if (iff->in(0) != n_ctrl) {
1228         return; // Compare must be in same blk as if
1229       }
1230     } else if (iff->is_CMove()) { // Trying to split-up a CMOVE
1231       // Can't split CMove with different control edge.
1232       if (iff->in(0) != NULL && iff->in(0) != n_ctrl ) {
1233         return;
1234       }
1235       if (get_ctrl(iff->in(2)) == n_ctrl ||
1236           get_ctrl(iff->in(3)) == n_ctrl) {
1237         return;                 // Inputs not yet split-up
1238       }
1239       if (get_loop(n_ctrl) != get_loop(get_ctrl(iff))) {
1240         return;                 // Loop-invar test gates loop-varying CMOVE
1241       }
1242     } else {
1243       return;  // some other kind of node, such as an Allocate
1244     }
1245 
1246     // When is split-if profitable?  Every 'win' on means some control flow
1247     // goes dead, so it's almost always a win.
1248     int policy = 0;
1249     // Split compare 'n' through the merge point if it is profitable
1250     Node *phi = split_thru_phi( n, n_ctrl, policy);
1251     if (!phi) {
1252       return;
1253     }
1254 
1255     // Found a Phi to split thru!
1256     // Replace 'n' with the new phi
1257     _igvn.replace_node(n, phi);
1258 
1259     // Now split the bool up thru the phi
1260     Node *bolphi = split_thru_phi(bol, n_ctrl, -1);
1261     guarantee(bolphi != NULL, "null boolean phi node");
1262 
1263     _igvn.replace_node(bol, bolphi);
1264     assert(iff->in(1) == bolphi, "");
1265 
1266     if (bolphi->Value(&_igvn)->singleton()) {
1267       return;
1268     }
1269 
1270     // Conditional-move?  Must split up now
1271     if (!iff->is_If()) {
1272       Node *cmovphi = split_thru_phi(iff, n_ctrl, -1);
1273       _igvn.replace_node(iff, cmovphi);
1274       return;
1275     }
1276 
1277     // Now split the IF
1278     do_split_if(iff);
1279     return;
1280   }
1281 
1282   // Two identical ifs back to back can be merged
1283   if (identical_backtoback_ifs(n) && can_split_if(n->in(0))) {
1284     Node *n_ctrl = n->in(0);
1285     PhiNode* bolphi = PhiNode::make_blank(n_ctrl, n->in(1));
1286     IfNode* dom_if = idom(n_ctrl)->as_If();
1287     Node* proj_true = dom_if->proj_out(1);
1288     Node* proj_false = dom_if->proj_out(0);
1289     Node* con_true = _igvn.makecon(TypeInt::ONE);
1290     Node* con_false = _igvn.makecon(TypeInt::ZERO);
1291 
1292     for (uint i = 1; i < n_ctrl->req(); i++) {
1293       if (is_dominator(proj_true, n_ctrl->in(i))) {
1294         bolphi->init_req(i, con_true);
1295       } else {
1296         assert(is_dominator(proj_false, n_ctrl->in(i)), "bad if");
1297         bolphi->init_req(i, con_false);
1298       }
1299     }
1300     register_new_node(bolphi, n_ctrl);
1301     _igvn.replace_input_of(n, 1, bolphi);
1302 
1303     // Now split the IF
1304     do_split_if(n);
1305     return;
1306   }
1307 
1308   // Check for an IF ready to split; one that has its
1309   // condition codes input coming from a Phi at the block start.
1310   int n_op = n->Opcode();
1311 
1312   // Check for an IF being dominated by another IF same test
1313   if (n_op == Op_If ||
1314       n_op == Op_RangeCheck) {
1315     Node *bol = n->in(1);
1316     uint max = bol->outcnt();
1317     // Check for same test used more than once?
1318     if (max > 1 && bol->is_Bool()) {
1319       // Search up IDOMs to see if this IF is dominated.
1320       Node *cutoff = get_ctrl(bol);
1321 
1322       // Now search up IDOMs till cutoff, looking for a dominating test
1323       Node *prevdom = n;
1324       Node *dom = idom(prevdom);
1325       while (dom != cutoff) {
1326         if (dom->req() > 1 && dom->in(1) == bol && prevdom->in(0) == dom) {
1327           // Replace the dominated test with an obvious true or false.
1328           // Place it on the IGVN worklist for later cleanup.
1329           C->set_major_progress();
1330           dominated_by(prevdom, n, false, true);
1331 #ifndef PRODUCT
1332           if( VerifyLoopOptimizations ) verify();
1333 #endif
1334           return;
1335         }
1336         prevdom = dom;
1337         dom = idom(prevdom);
1338       }
1339     }
1340   }
1341 
1342   // See if a shared loop-varying computation has no loop-varying uses.
1343   // Happens if something is only used for JVM state in uncommon trap exits,
1344   // like various versions of induction variable+offset.  Clone the
1345   // computation per usage to allow it to sink out of the loop.
1346   if (has_ctrl(n) && !n->in(0)) {// n not dead and has no control edge (can float about)
1347     Node *n_ctrl = get_ctrl(n);
1348     IdealLoopTree *n_loop = get_loop(n_ctrl);
1349     if( n_loop != _ltree_root ) {
1350       DUIterator_Fast imax, i = n->fast_outs(imax);
1351       for (; i < imax; i++) {
1352         Node* u = n->fast_out(i);
1353         if( !has_ctrl(u) )     break; // Found control user
1354         IdealLoopTree *u_loop = get_loop(get_ctrl(u));
1355         if( u_loop == n_loop ) break; // Found loop-varying use
1356         if( n_loop->is_member( u_loop ) ) break; // Found use in inner loop
1357         if( u->Opcode() == Op_Opaque1 ) break; // Found loop limit, bugfix for 4677003
1358       }
1359       bool did_break = (i < imax);  // Did we break out of the previous loop?
1360       if (!did_break && n->outcnt() > 1) { // All uses in outer loops!
1361         Node *late_load_ctrl = NULL;
1362         if (n->is_Load()) {
1363           // If n is a load, get and save the result from get_late_ctrl(),
1364           // to be later used in calculating the control for n's clones.
1365           clear_dom_lca_tags();
1366           late_load_ctrl = get_late_ctrl(n, n_ctrl);
1367         }
1368         // If n is a load, and the late control is the same as the current
1369         // control, then the cloning of n is a pointless exercise, because
1370         // GVN will ensure that we end up where we started.
1371         if (!n->is_Load() || late_load_ctrl != n_ctrl) {
1372           BarrierSetC2* bs = BarrierSet::barrier_set()->barrier_set_c2();
1373           for (DUIterator_Last jmin, j = n->last_outs(jmin); j >= jmin; ) {
1374             Node *u = n->last_out(j); // Clone private computation per use
1375             _igvn.rehash_node_delayed(u);
1376             Node *x = n->clone(); // Clone computation
1377             Node *x_ctrl = NULL;
1378             if( u->is_Phi() ) {
1379               // Replace all uses of normal nodes.  Replace Phi uses
1380               // individually, so the separate Nodes can sink down
1381               // different paths.
1382               uint k = 1;
1383               while( u->in(k) != n ) k++;
1384               u->set_req( k, x );
1385               // x goes next to Phi input path
1386               x_ctrl = u->in(0)->in(k);
1387               --j;
1388             } else {              // Normal use
1389               // Replace all uses
1390               for( uint k = 0; k < u->req(); k++ ) {
1391                 if( u->in(k) == n ) {
1392                   u->set_req( k, x );
1393                   --j;
1394                 }
1395               }
1396               x_ctrl = get_ctrl(u);
1397             }
1398 
1399             // Find control for 'x' next to use but not inside inner loops.
1400             // For inner loop uses get the preheader area.
1401             x_ctrl = place_near_use(x_ctrl);
1402 
1403             if (bs->sink_node(this, n, x, x_ctrl, n_ctrl)) {
1404               continue;
1405             }
1406 
1407             if (n->is_Load()) {
1408               // For loads, add a control edge to a CFG node outside of the loop
1409               // to force them to not combine and return back inside the loop
1410               // during GVN optimization (4641526).
1411               //
1412               // Because we are setting the actual control input, factor in
1413               // the result from get_late_ctrl() so we respect any
1414               // anti-dependences. (6233005).
1415               x_ctrl = dom_lca(late_load_ctrl, x_ctrl);
1416 
1417               // Don't allow the control input to be a CFG splitting node.
1418               // Such nodes should only have ProjNodes as outs, e.g. IfNode
1419               // should only have IfTrueNode and IfFalseNode (4985384).
1420               x_ctrl = find_non_split_ctrl(x_ctrl);
1421               
1422               IdealLoopTree* x_loop = get_loop(x_ctrl);
1423               if (x_loop->_head->is_OuterStripMinedLoop() && is_dominator(n_ctrl, x_loop->_head)) {
1424                 // Anti dependence analysis is sometimes too
1425                 // conservative: a store in the outer strip mined loop
1426                 // can prevent a load from floating out of the outer
1427                 // strip mined loop but the load may not be referenced
1428                 // from the safepoint: loop strip mining verification
1429                 // code reports a problem in that case. Make sure the
1430                 // load is not moved in the outer strip mined loop in
1431                 // that case.
1432                 x_ctrl = x_loop->_head->in(LoopNode::EntryControl);
1433               }
1434               assert(dom_depth(n_ctrl) <= dom_depth(x_ctrl), "n is later than its clone");
1435 
1436               x->set_req(0, x_ctrl);
1437             }
1438             register_new_node(x, x_ctrl);
1439 
1440             // Some institutional knowledge is needed here: 'x' is
1441             // yanked because if the optimizer runs GVN on it all the
1442             // cloned x's will common up and undo this optimization and
1443             // be forced back in the loop.
1444             // I tried setting control edges on the x's to force them to
1445             // not combine, but the matching gets worried when it tries
1446             // to fold a StoreP and an AddP together (as part of an
1447             // address expression) and the AddP and StoreP have
1448             // different controls.
1449             if (!x->is_Load() && !x->is_DecodeNarrowPtr()) _igvn._worklist.yank(x);
1450           }
1451           _igvn.remove_dead_node(n);
1452         }
1453       }
1454     }
1455   }
1456 
1457   try_move_store_after_loop(n);
1458 
1459   // Check for Opaque2's who's loop has disappeared - who's input is in the
1460   // same loop nest as their output.  Remove 'em, they are no longer useful.
1461   if( n_op == Op_Opaque2 &&
1462       n->in(1) != NULL &&
1463       get_loop(get_ctrl(n)) == get_loop(get_ctrl(n->in(1))) ) {
1464     _igvn.replace_node( n, n->in(1) );
1465   }
1466 }
1467 
1468 //------------------------------split_if_with_blocks---------------------------
1469 // Check for aggressive application of 'split-if' optimization,
1470 // using basic block level info.
1471 void PhaseIdealLoop::split_if_with_blocks(VectorSet &visited, Node_Stack &nstack) {
1472   Node* root = C->root();
1473   visited.set(root->_idx); // first, mark root as visited
1474   // Do pre-visit work for root
1475   Node* n   = split_if_with_blocks_pre(root);
1476   uint  cnt = n->outcnt();
1477   uint  i   = 0;
1478 
1479   while (true) {
1480     // Visit all children
1481     if (i < cnt) {
1482       Node* use = n->raw_out(i);
1483       ++i;
1484       if (use->outcnt() != 0 && !visited.test_set(use->_idx)) {
1485         // Now do pre-visit work for this use
1486         use = split_if_with_blocks_pre(use);
1487         nstack.push(n, i); // Save parent and next use's index.
1488         n   = use;         // Process all children of current use.
1489         cnt = use->outcnt();
1490         i   = 0;
1491       }
1492     }
1493     else {
1494       // All of n's children have been processed, complete post-processing.
1495       if (cnt != 0 && !n->is_Con()) {
1496         assert(has_node(n), "no dead nodes");
1497         split_if_with_blocks_post(n);
1498       }
1499       if (must_throttle_split_if()) {
1500         nstack.clear();
1501       }
1502       if (nstack.is_empty()) {
1503         // Finished all nodes on stack.
1504         break;
1505       }
1506       // Get saved parent node and next use's index. Visit the rest of uses.
1507       n   = nstack.node();
1508       cnt = n->outcnt();
1509       i   = nstack.index();
1510       nstack.pop();
1511     }
1512   }
1513 }
1514 
1515 
1516 //=============================================================================
1517 //
1518 //                   C L O N E   A   L O O P   B O D Y
1519 //
1520 
1521 //------------------------------clone_iff--------------------------------------
1522 // Passed in a Phi merging (recursively) some nearly equivalent Bool/Cmps.
1523 // "Nearly" because all Nodes have been cloned from the original in the loop,
1524 // but the fall-in edges to the Cmp are different.  Clone bool/Cmp pairs
1525 // through the Phi recursively, and return a Bool.
1526 Node* PhaseIdealLoop::clone_iff(PhiNode *phi, IdealLoopTree *loop) {
1527 
1528   // Convert this Phi into a Phi merging Bools
1529   uint i;
1530   for (i = 1; i < phi->req(); i++) {
1531     Node *b = phi->in(i);
1532     if (b->is_Phi()) {
1533       _igvn.replace_input_of(phi, i, clone_iff(b->as_Phi(), loop));
1534     } else {
1535       assert(b->is_Bool() || b->Opcode() == Op_Opaque4, "");
1536     }
1537   }
1538 
1539   Node* n = phi->in(1);
1540   Node* sample_opaque = NULL;
1541   Node *sample_bool = NULL;
1542   if (n->Opcode() == Op_Opaque4) {
1543     sample_opaque = n;
1544     sample_bool = n->in(1);
1545     assert(sample_bool->is_Bool(), "wrong type");
1546   } else {
1547     sample_bool = n;
1548   }
1549   Node *sample_cmp = sample_bool->in(1);
1550 
1551   // Make Phis to merge the Cmp's inputs.
1552   PhiNode *phi1 = new PhiNode(phi->in(0), Type::TOP);
1553   PhiNode *phi2 = new PhiNode(phi->in(0), Type::TOP);
1554   for (i = 1; i < phi->req(); i++) {
1555     Node *n1 = sample_opaque == NULL ? phi->in(i)->in(1)->in(1) : phi->in(i)->in(1)->in(1)->in(1);
1556     Node *n2 = sample_opaque == NULL ? phi->in(i)->in(1)->in(2) : phi->in(i)->in(1)->in(1)->in(2);
1557     phi1->set_req(i, n1);
1558     phi2->set_req(i, n2);
1559     phi1->set_type(phi1->type()->meet_speculative(n1->bottom_type()));
1560     phi2->set_type(phi2->type()->meet_speculative(n2->bottom_type()));
1561   }
1562   // See if these Phis have been made before.
1563   // Register with optimizer
1564   Node *hit1 = _igvn.hash_find_insert(phi1);
1565   if (hit1) {                   // Hit, toss just made Phi
1566     _igvn.remove_dead_node(phi1); // Remove new phi
1567     assert(hit1->is_Phi(), "" );
1568     phi1 = (PhiNode*)hit1;      // Use existing phi
1569   } else {                      // Miss
1570     _igvn.register_new_node_with_optimizer(phi1);
1571   }
1572   Node *hit2 = _igvn.hash_find_insert(phi2);
1573   if (hit2) {                   // Hit, toss just made Phi
1574     _igvn.remove_dead_node(phi2); // Remove new phi
1575     assert(hit2->is_Phi(), "" );
1576     phi2 = (PhiNode*)hit2;      // Use existing phi
1577   } else {                      // Miss
1578     _igvn.register_new_node_with_optimizer(phi2);
1579   }
1580   // Register Phis with loop/block info
1581   set_ctrl(phi1, phi->in(0));
1582   set_ctrl(phi2, phi->in(0));
1583   // Make a new Cmp
1584   Node *cmp = sample_cmp->clone();
1585   cmp->set_req(1, phi1);
1586   cmp->set_req(2, phi2);
1587   _igvn.register_new_node_with_optimizer(cmp);
1588   set_ctrl(cmp, phi->in(0));
1589 
1590   // Make a new Bool
1591   Node *b = sample_bool->clone();
1592   b->set_req(1,cmp);
1593   _igvn.register_new_node_with_optimizer(b);
1594   set_ctrl(b, phi->in(0));
1595 
1596   if (sample_opaque != NULL) {
1597     Node* opaque = sample_opaque->clone();
1598     opaque->set_req(1, b);
1599     _igvn.register_new_node_with_optimizer(opaque);
1600     set_ctrl(opaque, phi->in(0));
1601     return opaque;
1602   }
1603 
1604   assert(b->is_Bool(), "");
1605   return b;
1606 }
1607 
1608 //------------------------------clone_bool-------------------------------------
1609 // Passed in a Phi merging (recursively) some nearly equivalent Bool/Cmps.
1610 // "Nearly" because all Nodes have been cloned from the original in the loop,
1611 // but the fall-in edges to the Cmp are different.  Clone bool/Cmp pairs
1612 // through the Phi recursively, and return a Bool.
1613 CmpNode *PhaseIdealLoop::clone_bool( PhiNode *phi, IdealLoopTree *loop ) {
1614   uint i;
1615   // Convert this Phi into a Phi merging Bools
1616   for( i = 1; i < phi->req(); i++ ) {
1617     Node *b = phi->in(i);
1618     if( b->is_Phi() ) {
1619       _igvn.replace_input_of(phi, i, clone_bool( b->as_Phi(), loop ));
1620     } else {
1621       assert( b->is_Cmp() || b->is_top(), "inputs are all Cmp or TOP" );
1622     }
1623   }
1624 
1625   Node *sample_cmp = phi->in(1);
1626 
1627   // Make Phis to merge the Cmp's inputs.
1628   PhiNode *phi1 = new PhiNode( phi->in(0), Type::TOP );
1629   PhiNode *phi2 = new PhiNode( phi->in(0), Type::TOP );
1630   for( uint j = 1; j < phi->req(); j++ ) {
1631     Node *cmp_top = phi->in(j); // Inputs are all Cmp or TOP
1632     Node *n1, *n2;
1633     if( cmp_top->is_Cmp() ) {
1634       n1 = cmp_top->in(1);
1635       n2 = cmp_top->in(2);
1636     } else {
1637       n1 = n2 = cmp_top;
1638     }
1639     phi1->set_req( j, n1 );
1640     phi2->set_req( j, n2 );
1641     phi1->set_type(phi1->type()->meet_speculative(n1->bottom_type()));
1642     phi2->set_type(phi2->type()->meet_speculative(n2->bottom_type()));
1643   }
1644 
1645   // See if these Phis have been made before.
1646   // Register with optimizer
1647   Node *hit1 = _igvn.hash_find_insert(phi1);
1648   if( hit1 ) {                  // Hit, toss just made Phi
1649     _igvn.remove_dead_node(phi1); // Remove new phi
1650     assert( hit1->is_Phi(), "" );
1651     phi1 = (PhiNode*)hit1;      // Use existing phi
1652   } else {                      // Miss
1653     _igvn.register_new_node_with_optimizer(phi1);
1654   }
1655   Node *hit2 = _igvn.hash_find_insert(phi2);
1656   if( hit2 ) {                  // Hit, toss just made Phi
1657     _igvn.remove_dead_node(phi2); // Remove new phi
1658     assert( hit2->is_Phi(), "" );
1659     phi2 = (PhiNode*)hit2;      // Use existing phi
1660   } else {                      // Miss
1661     _igvn.register_new_node_with_optimizer(phi2);
1662   }
1663   // Register Phis with loop/block info
1664   set_ctrl(phi1, phi->in(0));
1665   set_ctrl(phi2, phi->in(0));
1666   // Make a new Cmp
1667   Node *cmp = sample_cmp->clone();
1668   cmp->set_req( 1, phi1 );
1669   cmp->set_req( 2, phi2 );
1670   _igvn.register_new_node_with_optimizer(cmp);
1671   set_ctrl(cmp, phi->in(0));
1672 
1673   assert( cmp->is_Cmp(), "" );
1674   return (CmpNode*)cmp;
1675 }
1676 
1677 //------------------------------sink_use---------------------------------------
1678 // If 'use' was in the loop-exit block, it now needs to be sunk
1679 // below the post-loop merge point.
1680 void PhaseIdealLoop::sink_use( Node *use, Node *post_loop ) {
1681   if (!use->is_CFG() && get_ctrl(use) == post_loop->in(2)) {
1682     set_ctrl(use, post_loop);
1683     for (DUIterator j = use->outs(); use->has_out(j); j++)
1684       sink_use(use->out(j), post_loop);
1685   }
1686 }
1687 
1688 void PhaseIdealLoop::clone_loop_handle_data_uses(Node* old, Node_List &old_new,
1689                                                  IdealLoopTree* loop, IdealLoopTree* outer_loop,
1690                                                  Node_List*& split_if_set, Node_List*& split_bool_set,
1691                                                  Node_List*& split_cex_set, Node_List& worklist,
1692                                                  uint new_counter, CloneLoopMode mode) {
1693   Node* nnn = old_new[old->_idx];
1694   // Copy uses to a worklist, so I can munge the def-use info
1695   // with impunity.
1696   for (DUIterator_Fast jmax, j = old->fast_outs(jmax); j < jmax; j++)
1697     worklist.push(old->fast_out(j));
1698 
1699   while( worklist.size() ) {
1700     Node *use = worklist.pop();
1701     if (!has_node(use))  continue; // Ignore dead nodes
1702     if (use->in(0) == C->top())  continue;
1703     IdealLoopTree *use_loop = get_loop( has_ctrl(use) ? get_ctrl(use) : use );
1704     // Check for data-use outside of loop - at least one of OLD or USE
1705     // must not be a CFG node.
1706 #ifdef ASSERT
1707     if (loop->_head->as_Loop()->is_strip_mined() && outer_loop->is_member(use_loop) && !loop->is_member(use_loop) && old_new[use->_idx] == NULL) {
1708       Node* sfpt = loop->_head->as_CountedLoop()->outer_safepoint();
1709       assert(mode == ControlAroundStripMined && use == sfpt, "missed a node");
1710     }
1711 #endif
1712     if (!loop->is_member(use_loop) && !outer_loop->is_member(use_loop) && (!old->is_CFG() || !use->is_CFG())) {
1713 
1714       // If the Data use is an IF, that means we have an IF outside of the
1715       // loop that is switching on a condition that is set inside of the
1716       // loop.  Happens if people set a loop-exit flag; then test the flag
1717       // in the loop to break the loop, then test is again outside of the
1718       // loop to determine which way the loop exited.
1719       // Loop predicate If node connects to Bool node through Opaque1 node.
1720       if (use->is_If() || use->is_CMove() || C->is_predicate_opaq(use) || use->Opcode() == Op_Opaque4) {
1721         // Since this code is highly unlikely, we lazily build the worklist
1722         // of such Nodes to go split.
1723         if (!split_if_set) {
1724           ResourceArea *area = Thread::current()->resource_area();
1725           split_if_set = new Node_List(area);
1726         }
1727         split_if_set->push(use);
1728       }
1729       if (use->is_Bool()) {
1730         if (!split_bool_set) {
1731           ResourceArea *area = Thread::current()->resource_area();
1732           split_bool_set = new Node_List(area);
1733         }
1734         split_bool_set->push(use);
1735       }
1736       if (use->Opcode() == Op_CreateEx) {
1737         if (!split_cex_set) {
1738           ResourceArea *area = Thread::current()->resource_area();
1739           split_cex_set = new Node_List(area);
1740         }
1741         split_cex_set->push(use);
1742       }
1743 
1744 
1745       // Get "block" use is in
1746       uint idx = 0;
1747       while( use->in(idx) != old ) idx++;
1748       Node *prev = use->is_CFG() ? use : get_ctrl(use);
1749       assert(!loop->is_member(get_loop(prev)) && !outer_loop->is_member(get_loop(prev)), "" );
1750       Node *cfg = prev->_idx >= new_counter
1751         ? prev->in(2)
1752         : idom(prev);
1753       if( use->is_Phi() )     // Phi use is in prior block
1754         cfg = prev->in(idx);  // NOT in block of Phi itself
1755       if (cfg->is_top()) {    // Use is dead?
1756         _igvn.replace_input_of(use, idx, C->top());
1757         continue;
1758       }
1759 
1760       // If use is referenced through control edge... (idx == 0)
1761       if (mode == IgnoreStripMined && idx == 0) {
1762         LoopNode *head = loop->_head->as_Loop();
1763         if (head->is_strip_mined() && is_dominator(head->outer_loop_exit(), prev)) {
1764           // That node is outside the inner loop, leave it outside the
1765           // outer loop as well to not confuse verification code.
1766           assert(!loop->_parent->is_member(use_loop), "should be out of the outer loop");
1767           _igvn.replace_input_of(use, 0, head->outer_loop_exit());
1768           continue;
1769         }
1770       }
1771 
1772       while(!outer_loop->is_member(get_loop(cfg))) {
1773         prev = cfg;
1774         cfg = cfg->_idx >= new_counter ? cfg->in(2) : idom(cfg);
1775       }
1776       // If the use occurs after merging several exits from the loop, then
1777       // old value must have dominated all those exits.  Since the same old
1778       // value was used on all those exits we did not need a Phi at this
1779       // merge point.  NOW we do need a Phi here.  Each loop exit value
1780       // is now merged with the peeled body exit; each exit gets its own
1781       // private Phi and those Phis need to be merged here.
1782       Node *phi;
1783       if( prev->is_Region() ) {
1784         if( idx == 0 ) {      // Updating control edge?
1785           phi = prev;         // Just use existing control
1786         } else {              // Else need a new Phi
1787           phi = PhiNode::make( prev, old );
1788           // Now recursively fix up the new uses of old!
1789           for( uint i = 1; i < prev->req(); i++ ) {
1790             worklist.push(phi); // Onto worklist once for each 'old' input
1791           }
1792         }
1793       } else {
1794         // Get new RegionNode merging old and new loop exits
1795         prev = old_new[prev->_idx];
1796         assert( prev, "just made this in step 7" );
1797         if( idx == 0) {      // Updating control edge?
1798           phi = prev;         // Just use existing control
1799         } else {              // Else need a new Phi
1800           // Make a new Phi merging data values properly
1801           phi = PhiNode::make( prev, old );
1802           phi->set_req( 1, nnn );
1803         }
1804       }
1805       // If inserting a new Phi, check for prior hits
1806       if( idx != 0 ) {
1807         Node *hit = _igvn.hash_find_insert(phi);
1808         if( hit == NULL ) {
1809           _igvn.register_new_node_with_optimizer(phi); // Register new phi
1810         } else {                                      // or
1811           // Remove the new phi from the graph and use the hit
1812           _igvn.remove_dead_node(phi);
1813           phi = hit;                                  // Use existing phi
1814         }
1815         set_ctrl(phi, prev);
1816       }
1817       // Make 'use' use the Phi instead of the old loop body exit value
1818       _igvn.replace_input_of(use, idx, phi);
1819       if( use->_idx >= new_counter ) { // If updating new phis
1820         // Not needed for correctness, but prevents a weak assert
1821         // in AddPNode from tripping (when we end up with different
1822         // base & derived Phis that will become the same after
1823         // IGVN does CSE).
1824         Node *hit = _igvn.hash_find_insert(use);
1825         if( hit )             // Go ahead and re-hash for hits.
1826           _igvn.replace_node( use, hit );
1827       }
1828 
1829       // If 'use' was in the loop-exit block, it now needs to be sunk
1830       // below the post-loop merge point.
1831       sink_use( use, prev );
1832     }
1833   }
1834 }
1835 
1836 static void clone_outer_loop_helper(Node* n, const IdealLoopTree *loop, const IdealLoopTree* outer_loop,
1837                                     const Node_List &old_new, Unique_Node_List& wq, PhaseIdealLoop* phase,
1838                                     bool check_old_new) {
1839   for (DUIterator_Fast jmax, j = n->fast_outs(jmax); j < jmax; j++) {
1840     Node* u = n->fast_out(j);
1841     assert(check_old_new || old_new[u->_idx] == NULL, "shouldn't have been cloned");
1842     if (!u->is_CFG() && (!check_old_new || old_new[u->_idx] == NULL)) {
1843       Node* c = phase->get_ctrl(u);
1844       IdealLoopTree* u_loop = phase->get_loop(c);
1845       assert(!loop->is_member(u_loop), "can be in outer loop or out of both loops only");
1846       if (outer_loop->is_member(u_loop)) {
1847         wq.push(u);
1848       }
1849     }
1850   }
1851 }
1852 
1853 void PhaseIdealLoop::clone_outer_loop(LoopNode* head, CloneLoopMode mode, IdealLoopTree *loop,
1854                                       IdealLoopTree* outer_loop, int dd, Node_List &old_new,
1855                                       Node_List& extra_data_nodes) {
1856   if (head->is_strip_mined() && mode != IgnoreStripMined) {
1857     CountedLoopNode* cl = head->as_CountedLoop();
1858     Node* l = cl->outer_loop();
1859     Node* tail = cl->outer_loop_tail();
1860     IfNode* le = cl->outer_loop_end();
1861     Node* sfpt = cl->outer_safepoint();
1862     CountedLoopEndNode* cle = cl->loopexit();
1863     CountedLoopNode* new_cl = old_new[cl->_idx]->as_CountedLoop();
1864     CountedLoopEndNode* new_cle = new_cl->as_CountedLoop()->loopexit_or_null();
1865     Node* cle_out = cle->proj_out(false);
1866 
1867     Node* new_sfpt = NULL;
1868     Node* new_cle_out = cle_out->clone();
1869     old_new.map(cle_out->_idx, new_cle_out);
1870     if (mode == CloneIncludesStripMined) {
1871       // clone outer loop body
1872       Node* new_l = l->clone();
1873       Node* new_tail = tail->clone();
1874       IfNode* new_le = le->clone()->as_If();
1875       new_sfpt = sfpt->clone();
1876 
1877       set_loop(new_l, outer_loop->_parent);
1878       set_idom(new_l, new_l->in(LoopNode::EntryControl), dd);
1879       set_loop(new_cle_out, outer_loop->_parent);
1880       set_idom(new_cle_out, new_cle, dd);
1881       set_loop(new_sfpt, outer_loop->_parent);
1882       set_idom(new_sfpt, new_cle_out, dd);
1883       set_loop(new_le, outer_loop->_parent);
1884       set_idom(new_le, new_sfpt, dd);
1885       set_loop(new_tail, outer_loop->_parent);
1886       set_idom(new_tail, new_le, dd);
1887       set_idom(new_cl, new_l, dd);
1888 
1889       old_new.map(l->_idx, new_l);
1890       old_new.map(tail->_idx, new_tail);
1891       old_new.map(le->_idx, new_le);
1892       old_new.map(sfpt->_idx, new_sfpt);
1893 
1894       new_l->set_req(LoopNode::LoopBackControl, new_tail);
1895       new_l->set_req(0, new_l);
1896       new_tail->set_req(0, new_le);
1897       new_le->set_req(0, new_sfpt);
1898       new_sfpt->set_req(0, new_cle_out);
1899       new_cle_out->set_req(0, new_cle);
1900       new_cl->set_req(LoopNode::EntryControl, new_l);
1901 
1902       _igvn.register_new_node_with_optimizer(new_l);
1903       _igvn.register_new_node_with_optimizer(new_tail);
1904       _igvn.register_new_node_with_optimizer(new_le);
1905     } else {
1906       Node *newhead = old_new[loop->_head->_idx];
1907       newhead->as_Loop()->clear_strip_mined();
1908       _igvn.replace_input_of(newhead, LoopNode::EntryControl, newhead->in(LoopNode::EntryControl)->in(LoopNode::EntryControl));
1909       set_idom(newhead, newhead->in(LoopNode::EntryControl), dd);
1910     }
1911     // Look at data node that were assigned a control in the outer
1912     // loop: they are kept in the outer loop by the safepoint so start
1913     // from the safepoint node's inputs.
1914     IdealLoopTree* outer_loop = get_loop(l);
1915     Node_Stack stack(2);
1916     stack.push(sfpt, 1);
1917     uint new_counter = C->unique();
1918     while (stack.size() > 0) {
1919       Node* n = stack.node();
1920       uint i = stack.index();
1921       while (i < n->req() &&
1922              (n->in(i) == NULL ||
1923               !has_ctrl(n->in(i)) ||
1924               get_loop(get_ctrl(n->in(i))) != outer_loop ||
1925               (old_new[n->in(i)->_idx] != NULL && old_new[n->in(i)->_idx]->_idx >= new_counter))) {
1926         i++;
1927       }
1928       if (i < n->req()) {
1929         stack.set_index(i+1);
1930         stack.push(n->in(i), 0);
1931       } else {
1932         assert(old_new[n->_idx] == NULL || n == sfpt || old_new[n->_idx]->_idx < new_counter, "no clone yet");
1933         Node* m = n == sfpt ? new_sfpt : n->clone();
1934         if (m != NULL) {
1935           for (uint i = 0; i < n->req(); i++) {
1936             if (m->in(i) != NULL && old_new[m->in(i)->_idx] != NULL) {
1937               m->set_req(i, old_new[m->in(i)->_idx]);
1938             }
1939           }
1940         } else {
1941           assert(n == sfpt && mode != CloneIncludesStripMined, "where's the safepoint clone?");
1942         }
1943         if (n != sfpt) {
1944           extra_data_nodes.push(n);
1945           _igvn.register_new_node_with_optimizer(m);
1946           assert(get_ctrl(n) == cle_out, "what other control?");
1947           set_ctrl(m, new_cle_out);
1948           old_new.map(n->_idx, m);
1949         }
1950         stack.pop();
1951       }
1952     }
1953     if (mode == CloneIncludesStripMined) {
1954       _igvn.register_new_node_with_optimizer(new_sfpt);
1955       _igvn.register_new_node_with_optimizer(new_cle_out);
1956     }
1957     // Some other transformation may have pessimistically assign some
1958     // data nodes to the outer loop. Set their control so they are out
1959     // of the outer loop.
1960     ResourceMark rm;
1961     Unique_Node_List wq;
1962     for (uint i = 0; i < extra_data_nodes.size(); i++) {
1963       Node* old = extra_data_nodes.at(i);
1964       clone_outer_loop_helper(old, loop, outer_loop, old_new, wq, this, true);
1965     }
1966     Node* new_ctrl = cl->outer_loop_exit();
1967     assert(get_loop(new_ctrl) != outer_loop, "must be out of the loop nest");
1968     for (uint i = 0; i < wq.size(); i++) {
1969       Node* n = wq.at(i);
1970       set_ctrl(n, new_ctrl);
1971       clone_outer_loop_helper(n, loop, outer_loop, old_new, wq, this, false);
1972     }
1973   } else {
1974     Node *newhead = old_new[loop->_head->_idx];
1975     set_idom(newhead, newhead->in(LoopNode::EntryControl), dd);
1976   }
1977 }
1978 
1979 //------------------------------clone_loop-------------------------------------
1980 //
1981 //                   C L O N E   A   L O O P   B O D Y
1982 //
1983 // This is the basic building block of the loop optimizations.  It clones an
1984 // entire loop body.  It makes an old_new loop body mapping; with this mapping
1985 // you can find the new-loop equivalent to an old-loop node.  All new-loop
1986 // nodes are exactly equal to their old-loop counterparts, all edges are the
1987 // same.  All exits from the old-loop now have a RegionNode that merges the
1988 // equivalent new-loop path.  This is true even for the normal "loop-exit"
1989 // condition.  All uses of loop-invariant old-loop values now come from (one
1990 // or more) Phis that merge their new-loop equivalents.
1991 //
1992 // This operation leaves the graph in an illegal state: there are two valid
1993 // control edges coming from the loop pre-header to both loop bodies.  I'll
1994 // definitely have to hack the graph after running this transform.
1995 //
1996 // From this building block I will further edit edges to perform loop peeling
1997 // or loop unrolling or iteration splitting (Range-Check-Elimination), etc.
1998 //
1999 // Parameter side_by_size_idom:
2000 //   When side_by_size_idom is NULL, the dominator tree is constructed for
2001 //      the clone loop to dominate the original.  Used in construction of
2002 //      pre-main-post loop sequence.
2003 //   When nonnull, the clone and original are side-by-side, both are
2004 //      dominated by the side_by_side_idom node.  Used in construction of
2005 //      unswitched loops.
2006 void PhaseIdealLoop::clone_loop( IdealLoopTree *loop, Node_List &old_new, int dd,
2007                                 CloneLoopMode mode, Node* side_by_side_idom) {
2008 
2009   LoopNode* head = loop->_head->as_Loop();
2010   head->verify_strip_mined(1);
2011 
2012   if (C->do_vector_loop() && PrintOpto) {
2013     const char* mname = C->method()->name()->as_quoted_ascii();
2014     if (mname != NULL) {
2015       tty->print("PhaseIdealLoop::clone_loop: for vectorize method %s\n", mname);
2016     }
2017   }
2018 
2019   CloneMap& cm = C->clone_map();
2020   Dict* dict = cm.dict();
2021   if (C->do_vector_loop()) {
2022     cm.set_clone_idx(cm.max_gen()+1);
2023 #ifndef PRODUCT
2024     if (PrintOpto) {
2025       tty->print_cr("PhaseIdealLoop::clone_loop: _clone_idx %d", cm.clone_idx());
2026       loop->dump_head();
2027     }
2028 #endif
2029   }
2030 
2031   // Step 1: Clone the loop body.  Make the old->new mapping.
2032   uint i;
2033   for( i = 0; i < loop->_body.size(); i++ ) {
2034     Node *old = loop->_body.at(i);
2035     Node *nnn = old->clone();
2036     old_new.map( old->_idx, nnn );
2037     if (C->do_vector_loop()) {
2038       cm.verify_insert_and_clone(old, nnn, cm.clone_idx());
2039     }
2040     _igvn.register_new_node_with_optimizer(nnn);
2041   }
2042 
2043   IdealLoopTree* outer_loop = (head->is_strip_mined() && mode != IgnoreStripMined) ? get_loop(head->as_CountedLoop()->outer_loop()) : loop;
2044 
2045   // Step 2: Fix the edges in the new body.  If the old input is outside the
2046   // loop use it.  If the old input is INside the loop, use the corresponding
2047   // new node instead.
2048   for( i = 0; i < loop->_body.size(); i++ ) {
2049     Node *old = loop->_body.at(i);
2050     Node *nnn = old_new[old->_idx];
2051     // Fix CFG/Loop controlling the new node
2052     if (has_ctrl(old)) {
2053       set_ctrl(nnn, old_new[get_ctrl(old)->_idx]);
2054     } else {
2055       set_loop(nnn, outer_loop->_parent);
2056       if (old->outcnt() > 0) {
2057         set_idom( nnn, old_new[idom(old)->_idx], dd );
2058       }
2059     }
2060     // Correct edges to the new node
2061     for( uint j = 0; j < nnn->req(); j++ ) {
2062         Node *n = nnn->in(j);
2063         if( n ) {
2064           IdealLoopTree *old_in_loop = get_loop( has_ctrl(n) ? get_ctrl(n) : n );
2065           if( loop->is_member( old_in_loop ) )
2066             nnn->set_req(j, old_new[n->_idx]);
2067         }
2068     }
2069     _igvn.hash_find_insert(nnn);
2070   }
2071 
2072   ResourceArea *area = Thread::current()->resource_area();
2073   Node_List extra_data_nodes(area); // data nodes in the outer strip mined loop
2074   clone_outer_loop(head, mode, loop, outer_loop, dd, old_new, extra_data_nodes);
2075 
2076   // Step 3: Now fix control uses.  Loop varying control uses have already
2077   // been fixed up (as part of all input edges in Step 2).  Loop invariant
2078   // control uses must be either an IfFalse or an IfTrue.  Make a merge
2079   // point to merge the old and new IfFalse/IfTrue nodes; make the use
2080   // refer to this.
2081   Node_List worklist(area);
2082   uint new_counter = C->unique();
2083   for( i = 0; i < loop->_body.size(); i++ ) {
2084     Node* old = loop->_body.at(i);
2085     if( !old->is_CFG() ) continue;
2086 
2087     // Copy uses to a worklist, so I can munge the def-use info
2088     // with impunity.
2089     for (DUIterator_Fast jmax, j = old->fast_outs(jmax); j < jmax; j++)
2090       worklist.push(old->fast_out(j));
2091 
2092     while( worklist.size() ) {  // Visit all uses
2093       Node *use = worklist.pop();
2094       if (!has_node(use))  continue; // Ignore dead nodes
2095       IdealLoopTree *use_loop = get_loop( has_ctrl(use) ? get_ctrl(use) : use );
2096       if( !loop->is_member( use_loop ) && use->is_CFG() ) {
2097         // Both OLD and USE are CFG nodes here.
2098         assert( use->is_Proj(), "" );
2099         Node* nnn = old_new[old->_idx];
2100 
2101         Node* newuse = NULL;
2102         if (head->is_strip_mined() && mode != IgnoreStripMined) {
2103           CountedLoopNode* cl = head->as_CountedLoop();
2104           CountedLoopEndNode* cle = cl->loopexit();
2105           Node* cle_out = cle->proj_out_or_null(false);
2106           if (use == cle_out) {
2107             IfNode* le = cl->outer_loop_end();
2108             use = le->proj_out(false);
2109             use_loop = get_loop(use);
2110             if (mode == CloneIncludesStripMined) {
2111               nnn = old_new[le->_idx];
2112             } else {
2113               newuse = old_new[cle_out->_idx];
2114             }
2115           }
2116         }
2117         if (newuse == NULL) {
2118           newuse = use->clone();
2119         }
2120 
2121         // Clone the loop exit control projection
2122         if (C->do_vector_loop()) {
2123           cm.verify_insert_and_clone(use, newuse, cm.clone_idx());
2124         }
2125         newuse->set_req(0,nnn);
2126         _igvn.register_new_node_with_optimizer(newuse);
2127         set_loop(newuse, use_loop);
2128         set_idom(newuse, nnn, dom_depth(nnn) + 1 );
2129 
2130         // We need a Region to merge the exit from the peeled body and the
2131         // exit from the old loop body.
2132         RegionNode *r = new RegionNode(3);
2133         // Map the old use to the new merge point
2134         old_new.map( use->_idx, r );
2135         uint dd_r = MIN2(dom_depth(newuse),dom_depth(use));
2136         assert( dd_r >= dom_depth(dom_lca(newuse,use)), "" );
2137 
2138         // The original user of 'use' uses 'r' instead.
2139         for (DUIterator_Last lmin, l = use->last_outs(lmin); l >= lmin;) {
2140           Node* useuse = use->last_out(l);
2141           _igvn.rehash_node_delayed(useuse);
2142           uint uses_found = 0;
2143           if( useuse->in(0) == use ) {
2144             useuse->set_req(0, r);
2145             uses_found++;
2146             if( useuse->is_CFG() ) {
2147               assert( dom_depth(useuse) > dd_r, "" );
2148               set_idom(useuse, r, dom_depth(useuse));
2149             }
2150           }
2151           for( uint k = 1; k < useuse->req(); k++ ) {
2152             if( useuse->in(k) == use ) {
2153               useuse->set_req(k, r);
2154               uses_found++;
2155               if (useuse->is_Loop() && k == LoopNode::EntryControl) {
2156                 assert(dom_depth(useuse) > dd_r , "");
2157                 set_idom(useuse, r, dom_depth(useuse));
2158               }
2159             }
2160           }
2161           l -= uses_found;    // we deleted 1 or more copies of this edge
2162         }
2163 
2164         // Now finish up 'r'
2165         r->set_req( 1, newuse );
2166         r->set_req( 2,    use );
2167         _igvn.register_new_node_with_optimizer(r);
2168         set_loop(r, use_loop);
2169         set_idom(r, !side_by_side_idom ? newuse->in(0) : side_by_side_idom, dd_r);
2170       } // End of if a loop-exit test
2171     }
2172   }
2173 
2174   // Step 4: If loop-invariant use is not control, it must be dominated by a
2175   // loop exit IfFalse/IfTrue.  Find "proper" loop exit.  Make a Region
2176   // there if needed.  Make a Phi there merging old and new used values.
2177   Node_List *split_if_set = NULL;
2178   Node_List *split_bool_set = NULL;
2179   Node_List *split_cex_set = NULL;
2180   for( i = 0; i < loop->_body.size(); i++ ) {
2181     Node* old = loop->_body.at(i);
2182     clone_loop_handle_data_uses(old, old_new, loop, outer_loop, split_if_set,
2183                                 split_bool_set, split_cex_set, worklist, new_counter,
2184                                 mode);
2185   }
2186 
2187   for (i = 0; i < extra_data_nodes.size(); i++) {
2188     Node* old = extra_data_nodes.at(i);
2189     clone_loop_handle_data_uses(old, old_new, loop, outer_loop, split_if_set,
2190                                 split_bool_set, split_cex_set, worklist, new_counter,
2191                                 mode);
2192   }
2193 
2194   // Check for IFs that need splitting/cloning.  Happens if an IF outside of
2195   // the loop uses a condition set in the loop.  The original IF probably
2196   // takes control from one or more OLD Regions (which in turn get from NEW
2197   // Regions).  In any case, there will be a set of Phis for each merge point
2198   // from the IF up to where the original BOOL def exists the loop.
2199   if (split_if_set) {
2200     while (split_if_set->size()) {
2201       Node *iff = split_if_set->pop();
2202       if (iff->in(1)->is_Phi()) {
2203         Node *b = clone_iff(iff->in(1)->as_Phi(), loop);
2204         _igvn.replace_input_of(iff, 1, b);
2205       }
2206     }
2207   }
2208   if (split_bool_set) {
2209     while (split_bool_set->size()) {
2210       Node *b = split_bool_set->pop();
2211       Node *phi = b->in(1);
2212       assert(phi->is_Phi(), "");
2213       CmpNode *cmp = clone_bool((PhiNode*)phi, loop);
2214       _igvn.replace_input_of(b, 1, cmp);
2215     }
2216   }
2217   if (split_cex_set) {
2218     while (split_cex_set->size()) {
2219       Node *b = split_cex_set->pop();
2220       assert(b->in(0)->is_Region(), "");
2221       assert(b->in(1)->is_Phi(), "");
2222       assert(b->in(0)->in(0) == b->in(1)->in(0), "");
2223       split_up(b, b->in(0), NULL);
2224     }
2225   }
2226 
2227 }
2228 
2229 
2230 //---------------------- stride_of_possible_iv -------------------------------------
2231 // Looks for an iff/bool/comp with one operand of the compare
2232 // being a cycle involving an add and a phi,
2233 // with an optional truncation (left-shift followed by a right-shift)
2234 // of the add. Returns zero if not an iv.
2235 int PhaseIdealLoop::stride_of_possible_iv(Node* iff) {
2236   Node* trunc1 = NULL;
2237   Node* trunc2 = NULL;
2238   const TypeInt* ttype = NULL;
2239   if (!iff->is_If() || iff->in(1) == NULL || !iff->in(1)->is_Bool()) {
2240     return 0;
2241   }
2242   BoolNode* bl = iff->in(1)->as_Bool();
2243   Node* cmp = bl->in(1);
2244   if (!cmp || (cmp->Opcode() != Op_CmpI && cmp->Opcode() != Op_CmpU)) {
2245     return 0;
2246   }
2247   // Must have an invariant operand
2248   if (is_member(get_loop(iff), get_ctrl(cmp->in(2)))) {
2249     return 0;
2250   }
2251   Node* add2 = NULL;
2252   Node* cmp1 = cmp->in(1);
2253   if (cmp1->is_Phi()) {
2254     // (If (Bool (CmpX phi:(Phi ...(Optional-trunc(AddI phi add2))) )))
2255     Node* phi = cmp1;
2256     for (uint i = 1; i < phi->req(); i++) {
2257       Node* in = phi->in(i);
2258       Node* add = CountedLoopNode::match_incr_with_optional_truncation(in,
2259                                 &trunc1, &trunc2, &ttype);
2260       if (add && add->in(1) == phi) {
2261         add2 = add->in(2);
2262         break;
2263       }
2264     }
2265   } else {
2266     // (If (Bool (CmpX addtrunc:(Optional-trunc((AddI (Phi ...addtrunc...) add2)) )))
2267     Node* addtrunc = cmp1;
2268     Node* add = CountedLoopNode::match_incr_with_optional_truncation(addtrunc,
2269                                 &trunc1, &trunc2, &ttype);
2270     if (add && add->in(1)->is_Phi()) {
2271       Node* phi = add->in(1);
2272       for (uint i = 1; i < phi->req(); i++) {
2273         if (phi->in(i) == addtrunc) {
2274           add2 = add->in(2);
2275           break;
2276         }
2277       }
2278     }
2279   }
2280   if (add2 != NULL) {
2281     const TypeInt* add2t = _igvn.type(add2)->is_int();
2282     if (add2t->is_con()) {
2283       return add2t->get_con();
2284     }
2285   }
2286   return 0;
2287 }
2288 
2289 
2290 //---------------------- stay_in_loop -------------------------------------
2291 // Return the (unique) control output node that's in the loop (if it exists.)
2292 Node* PhaseIdealLoop::stay_in_loop( Node* n, IdealLoopTree *loop) {
2293   Node* unique = NULL;
2294   if (!n) return NULL;
2295   for (DUIterator_Fast imax, i = n->fast_outs(imax); i < imax; i++) {
2296     Node* use = n->fast_out(i);
2297     if (!has_ctrl(use) && loop->is_member(get_loop(use))) {
2298       if (unique != NULL) {
2299         return NULL;
2300       }
2301       unique = use;
2302     }
2303   }
2304   return unique;
2305 }
2306 
2307 //------------------------------ register_node -------------------------------------
2308 // Utility to register node "n" with PhaseIdealLoop
2309 void PhaseIdealLoop::register_node(Node* n, IdealLoopTree *loop, Node* pred, int ddepth) {
2310   _igvn.register_new_node_with_optimizer(n);
2311   loop->_body.push(n);
2312   if (n->is_CFG()) {
2313     set_loop(n, loop);
2314     set_idom(n, pred, ddepth);
2315   } else {
2316     set_ctrl(n, pred);
2317   }
2318 }
2319 
2320 //------------------------------ proj_clone -------------------------------------
2321 // Utility to create an if-projection
2322 ProjNode* PhaseIdealLoop::proj_clone(ProjNode* p, IfNode* iff) {
2323   ProjNode* c = p->clone()->as_Proj();
2324   c->set_req(0, iff);
2325   return c;
2326 }
2327 
2328 //------------------------------ short_circuit_if -------------------------------------
2329 // Force the iff control output to be the live_proj
2330 Node* PhaseIdealLoop::short_circuit_if(IfNode* iff, ProjNode* live_proj) {
2331   guarantee(live_proj != NULL, "null projection");
2332   int proj_con = live_proj->_con;
2333   assert(proj_con == 0 || proj_con == 1, "false or true projection");
2334   Node *con = _igvn.intcon(proj_con);
2335   set_ctrl(con, C->root());
2336   if (iff) {
2337     iff->set_req(1, con);
2338   }
2339   return con;
2340 }
2341 
2342 //------------------------------ insert_if_before_proj -------------------------------------
2343 // Insert a new if before an if projection (* - new node)
2344 //
2345 // before
2346 //           if(test)
2347 //           /     \
2348 //          v       v
2349 //    other-proj   proj (arg)
2350 //
2351 // after
2352 //           if(test)
2353 //           /     \
2354 //          /       v
2355 //         |      * proj-clone
2356 //         v          |
2357 //    other-proj      v
2358 //                * new_if(relop(cmp[IU](left,right)))
2359 //                  /  \
2360 //                 v    v
2361 //         * new-proj  proj
2362 //         (returned)
2363 //
2364 ProjNode* PhaseIdealLoop::insert_if_before_proj(Node* left, bool Signed, BoolTest::mask relop, Node* right, ProjNode* proj) {
2365   IfNode* iff = proj->in(0)->as_If();
2366   IdealLoopTree *loop = get_loop(proj);
2367   ProjNode *other_proj = iff->proj_out(!proj->is_IfTrue())->as_Proj();
2368   int ddepth = dom_depth(proj);
2369 
2370   _igvn.rehash_node_delayed(iff);
2371   _igvn.rehash_node_delayed(proj);
2372 
2373   proj->set_req(0, NULL);  // temporary disconnect
2374   ProjNode* proj2 = proj_clone(proj, iff);
2375   register_node(proj2, loop, iff, ddepth);
2376 
2377   Node* cmp = Signed ? (Node*) new CmpINode(left, right) : (Node*) new CmpUNode(left, right);
2378   register_node(cmp, loop, proj2, ddepth);
2379 
2380   BoolNode* bol = new BoolNode(cmp, relop);
2381   register_node(bol, loop, proj2, ddepth);
2382 
2383   int opcode = iff->Opcode();
2384   assert(opcode == Op_If || opcode == Op_RangeCheck, "unexpected opcode");
2385   IfNode* new_if = (opcode == Op_If) ? new IfNode(proj2, bol, iff->_prob, iff->_fcnt):
2386     new RangeCheckNode(proj2, bol, iff->_prob, iff->_fcnt);
2387   register_node(new_if, loop, proj2, ddepth);
2388 
2389   proj->set_req(0, new_if); // reattach
2390   set_idom(proj, new_if, ddepth);
2391 
2392   ProjNode* new_exit = proj_clone(other_proj, new_if)->as_Proj();
2393   guarantee(new_exit != NULL, "null exit node");
2394   register_node(new_exit, get_loop(other_proj), new_if, ddepth);
2395 
2396   return new_exit;
2397 }
2398 
2399 //------------------------------ insert_region_before_proj -------------------------------------
2400 // Insert a region before an if projection (* - new node)
2401 //
2402 // before
2403 //           if(test)
2404 //          /      |
2405 //         v       |
2406 //       proj      v
2407 //               other-proj
2408 //
2409 // after
2410 //           if(test)
2411 //          /      |
2412 //         v       |
2413 // * proj-clone    v
2414 //         |     other-proj
2415 //         v
2416 // * new-region
2417 //         |
2418 //         v
2419 // *      dum_if
2420 //       /     \
2421 //      v       \
2422 // * dum-proj    v
2423 //              proj
2424 //
2425 RegionNode* PhaseIdealLoop::insert_region_before_proj(ProjNode* proj) {
2426   IfNode* iff = proj->in(0)->as_If();
2427   IdealLoopTree *loop = get_loop(proj);
2428   ProjNode *other_proj = iff->proj_out(!proj->is_IfTrue())->as_Proj();
2429   int ddepth = dom_depth(proj);
2430 
2431   _igvn.rehash_node_delayed(iff);
2432   _igvn.rehash_node_delayed(proj);
2433 
2434   proj->set_req(0, NULL);  // temporary disconnect
2435   ProjNode* proj2 = proj_clone(proj, iff);
2436   register_node(proj2, loop, iff, ddepth);
2437 
2438   RegionNode* reg = new RegionNode(2);
2439   reg->set_req(1, proj2);
2440   register_node(reg, loop, iff, ddepth);
2441 
2442   IfNode* dum_if = new IfNode(reg, short_circuit_if(NULL, proj), iff->_prob, iff->_fcnt);
2443   register_node(dum_if, loop, reg, ddepth);
2444 
2445   proj->set_req(0, dum_if); // reattach
2446   set_idom(proj, dum_if, ddepth);
2447 
2448   ProjNode* dum_proj = proj_clone(other_proj, dum_if);
2449   register_node(dum_proj, loop, dum_if, ddepth);
2450 
2451   return reg;
2452 }
2453 
2454 //------------------------------ insert_cmpi_loop_exit -------------------------------------
2455 // Clone a signed compare loop exit from an unsigned compare and
2456 // insert it before the unsigned cmp on the stay-in-loop path.
2457 // All new nodes inserted in the dominator tree between the original
2458 // if and it's projections.  The original if test is replaced with
2459 // a constant to force the stay-in-loop path.
2460 //
2461 // This is done to make sure that the original if and it's projections
2462 // still dominate the same set of control nodes, that the ctrl() relation
2463 // from data nodes to them is preserved, and that their loop nesting is
2464 // preserved.
2465 //
2466 // before
2467 //          if(i <u limit)    unsigned compare loop exit
2468 //         /       |
2469 //        v        v
2470 //   exit-proj   stay-in-loop-proj
2471 //
2472 // after
2473 //          if(stay-in-loop-const)  original if
2474 //         /       |
2475 //        /        v
2476 //       /  if(i <  limit)    new signed test
2477 //      /  /       |
2478 //     /  /        v
2479 //    /  /  if(i <u limit)    new cloned unsigned test
2480 //   /  /   /      |
2481 //   v  v  v       |
2482 //    region       |
2483 //        |        |
2484 //      dum-if     |
2485 //     /  |        |
2486 // ether  |        |
2487 //        v        v
2488 //   exit-proj   stay-in-loop-proj
2489 //
2490 IfNode* PhaseIdealLoop::insert_cmpi_loop_exit(IfNode* if_cmpu, IdealLoopTree *loop) {
2491   const bool Signed   = true;
2492   const bool Unsigned = false;
2493 
2494   BoolNode* bol = if_cmpu->in(1)->as_Bool();
2495   if (bol->_test._test != BoolTest::lt) return NULL;
2496   CmpNode* cmpu = bol->in(1)->as_Cmp();
2497   if (cmpu->Opcode() != Op_CmpU) return NULL;
2498   int stride = stride_of_possible_iv(if_cmpu);
2499   if (stride == 0) return NULL;
2500 
2501   Node* lp_proj = stay_in_loop(if_cmpu, loop);
2502   guarantee(lp_proj != NULL, "null loop node");
2503 
2504   ProjNode* lp_continue = lp_proj->as_Proj();
2505   ProjNode* lp_exit     = if_cmpu->proj_out(!lp_continue->is_IfTrue())->as_Proj();
2506 
2507   Node* limit = NULL;
2508   if (stride > 0) {
2509     limit = cmpu->in(2);
2510   } else {
2511     limit = _igvn.makecon(TypeInt::ZERO);
2512     set_ctrl(limit, C->root());
2513   }
2514   // Create a new region on the exit path
2515   RegionNode* reg = insert_region_before_proj(lp_exit);
2516   guarantee(reg != NULL, "null region node");
2517 
2518   // Clone the if-cmpu-true-false using a signed compare
2519   BoolTest::mask rel_i = stride > 0 ? bol->_test._test : BoolTest::ge;
2520   ProjNode* cmpi_exit = insert_if_before_proj(cmpu->in(1), Signed, rel_i, limit, lp_continue);
2521   reg->add_req(cmpi_exit);
2522 
2523   // Clone the if-cmpu-true-false
2524   BoolTest::mask rel_u = bol->_test._test;
2525   ProjNode* cmpu_exit = insert_if_before_proj(cmpu->in(1), Unsigned, rel_u, cmpu->in(2), lp_continue);
2526   reg->add_req(cmpu_exit);
2527 
2528   // Force original if to stay in loop.
2529   short_circuit_if(if_cmpu, lp_continue);
2530 
2531   return cmpi_exit->in(0)->as_If();
2532 }
2533 
2534 //------------------------------ remove_cmpi_loop_exit -------------------------------------
2535 // Remove a previously inserted signed compare loop exit.
2536 void PhaseIdealLoop::remove_cmpi_loop_exit(IfNode* if_cmp, IdealLoopTree *loop) {
2537   Node* lp_proj = stay_in_loop(if_cmp, loop);
2538   assert(if_cmp->in(1)->in(1)->Opcode() == Op_CmpI &&
2539          stay_in_loop(lp_proj, loop)->is_If() &&
2540          stay_in_loop(lp_proj, loop)->in(1)->in(1)->Opcode() == Op_CmpU, "inserted cmpi before cmpu");
2541   Node *con = _igvn.makecon(lp_proj->is_IfTrue() ? TypeInt::ONE : TypeInt::ZERO);
2542   set_ctrl(con, C->root());
2543   if_cmp->set_req(1, con);
2544 }
2545 
2546 //------------------------------ scheduled_nodelist -------------------------------------
2547 // Create a post order schedule of nodes that are in the
2548 // "member" set.  The list is returned in "sched".
2549 // The first node in "sched" is the loop head, followed by
2550 // nodes which have no inputs in the "member" set, and then
2551 // followed by the nodes that have an immediate input dependence
2552 // on a node in "sched".
2553 void PhaseIdealLoop::scheduled_nodelist( IdealLoopTree *loop, VectorSet& member, Node_List &sched ) {
2554 
2555   assert(member.test(loop->_head->_idx), "loop head must be in member set");
2556   Arena *a = Thread::current()->resource_area();
2557   VectorSet visited(a);
2558   Node_Stack nstack(a, loop->_body.size());
2559 
2560   Node* n  = loop->_head;  // top of stack is cached in "n"
2561   uint idx = 0;
2562   visited.set(n->_idx);
2563 
2564   // Initially push all with no inputs from within member set
2565   for(uint i = 0; i < loop->_body.size(); i++ ) {
2566     Node *elt = loop->_body.at(i);
2567     if (member.test(elt->_idx)) {
2568       bool found = false;
2569       for (uint j = 0; j < elt->req(); j++) {
2570         Node* def = elt->in(j);
2571         if (def && member.test(def->_idx) && def != elt) {
2572           found = true;
2573           break;
2574         }
2575       }
2576       if (!found && elt != loop->_head) {
2577         nstack.push(n, idx);
2578         n = elt;
2579         assert(!visited.test(n->_idx), "not seen yet");
2580         visited.set(n->_idx);
2581       }
2582     }
2583   }
2584 
2585   // traverse out's that are in the member set
2586   while (true) {
2587     if (idx < n->outcnt()) {
2588       Node* use = n->raw_out(idx);
2589       idx++;
2590       if (!visited.test_set(use->_idx)) {
2591         if (member.test(use->_idx)) {
2592           nstack.push(n, idx);
2593           n = use;
2594           idx = 0;
2595         }
2596       }
2597     } else {
2598       // All outputs processed
2599       sched.push(n);
2600       if (nstack.is_empty()) break;
2601       n   = nstack.node();
2602       idx = nstack.index();
2603       nstack.pop();
2604     }
2605   }
2606 }
2607 
2608 
2609 //------------------------------ has_use_in_set -------------------------------------
2610 // Has a use in the vector set
2611 bool PhaseIdealLoop::has_use_in_set( Node* n, VectorSet& vset ) {
2612   for (DUIterator_Fast jmax, j = n->fast_outs(jmax); j < jmax; j++) {
2613     Node* use = n->fast_out(j);
2614     if (vset.test(use->_idx)) {
2615       return true;
2616     }
2617   }
2618   return false;
2619 }
2620 
2621 
2622 //------------------------------ has_use_internal_to_set -------------------------------------
2623 // Has use internal to the vector set (ie. not in a phi at the loop head)
2624 bool PhaseIdealLoop::has_use_internal_to_set( Node* n, VectorSet& vset, IdealLoopTree *loop ) {
2625   Node* head  = loop->_head;
2626   for (DUIterator_Fast jmax, j = n->fast_outs(jmax); j < jmax; j++) {
2627     Node* use = n->fast_out(j);
2628     if (vset.test(use->_idx) && !(use->is_Phi() && use->in(0) == head)) {
2629       return true;
2630     }
2631   }
2632   return false;
2633 }
2634 
2635 
2636 //------------------------------ clone_for_use_outside_loop -------------------------------------
2637 // clone "n" for uses that are outside of loop
2638 int PhaseIdealLoop::clone_for_use_outside_loop( IdealLoopTree *loop, Node* n, Node_List& worklist ) {
2639   int cloned = 0;
2640   assert(worklist.size() == 0, "should be empty");
2641   for (DUIterator_Fast jmax, j = n->fast_outs(jmax); j < jmax; j++) {
2642     Node* use = n->fast_out(j);
2643     if( !loop->is_member(get_loop(has_ctrl(use) ? get_ctrl(use) : use)) ) {
2644       worklist.push(use);
2645     }
2646   }
2647   while( worklist.size() ) {
2648     Node *use = worklist.pop();
2649     if (!has_node(use) || use->in(0) == C->top()) continue;
2650     uint j;
2651     for (j = 0; j < use->req(); j++) {
2652       if (use->in(j) == n) break;
2653     }
2654     assert(j < use->req(), "must be there");
2655 
2656     // clone "n" and insert it between the inputs of "n" and the use outside the loop
2657     Node* n_clone = n->clone();
2658     _igvn.replace_input_of(use, j, n_clone);
2659     cloned++;
2660     Node* use_c;
2661     if (!use->is_Phi()) {
2662       use_c = has_ctrl(use) ? get_ctrl(use) : use->in(0);
2663     } else {
2664       // Use in a phi is considered a use in the associated predecessor block
2665       use_c = use->in(0)->in(j);
2666     }
2667     set_ctrl(n_clone, use_c);
2668     assert(!loop->is_member(get_loop(use_c)), "should be outside loop");
2669     get_loop(use_c)->_body.push(n_clone);
2670     _igvn.register_new_node_with_optimizer(n_clone);
2671 #ifndef PRODUCT
2672     if (TracePartialPeeling) {
2673       tty->print_cr("loop exit cloning old: %d new: %d newbb: %d", n->_idx, n_clone->_idx, get_ctrl(n_clone)->_idx);
2674     }
2675 #endif
2676   }
2677   return cloned;
2678 }
2679 
2680 
2681 //------------------------------ clone_for_special_use_inside_loop -------------------------------------
2682 // clone "n" for special uses that are in the not_peeled region.
2683 // If these def-uses occur in separate blocks, the code generator
2684 // marks the method as not compilable.  For example, if a "BoolNode"
2685 // is in a different basic block than the "IfNode" that uses it, then
2686 // the compilation is aborted in the code generator.
2687 void PhaseIdealLoop::clone_for_special_use_inside_loop( IdealLoopTree *loop, Node* n,
2688                                                         VectorSet& not_peel, Node_List& sink_list, Node_List& worklist ) {
2689   if (n->is_Phi() || n->is_Load()) {
2690     return;
2691   }
2692   assert(worklist.size() == 0, "should be empty");
2693   for (DUIterator_Fast jmax, j = n->fast_outs(jmax); j < jmax; j++) {
2694     Node* use = n->fast_out(j);
2695     if ( not_peel.test(use->_idx) &&
2696          (use->is_If() || use->is_CMove() || use->is_Bool()) &&
2697          use->in(1) == n)  {
2698       worklist.push(use);
2699     }
2700   }
2701   if (worklist.size() > 0) {
2702     // clone "n" and insert it between inputs of "n" and the use
2703     Node* n_clone = n->clone();
2704     loop->_body.push(n_clone);
2705     _igvn.register_new_node_with_optimizer(n_clone);
2706     set_ctrl(n_clone, get_ctrl(n));
2707     sink_list.push(n_clone);
2708     not_peel <<= n_clone->_idx;  // add n_clone to not_peel set.
2709 #ifndef PRODUCT
2710     if (TracePartialPeeling) {
2711       tty->print_cr("special not_peeled cloning old: %d new: %d", n->_idx, n_clone->_idx);
2712     }
2713 #endif
2714     while( worklist.size() ) {
2715       Node *use = worklist.pop();
2716       _igvn.rehash_node_delayed(use);
2717       for (uint j = 1; j < use->req(); j++) {
2718         if (use->in(j) == n) {
2719           use->set_req(j, n_clone);
2720         }
2721       }
2722     }
2723   }
2724 }
2725 
2726 
2727 //------------------------------ insert_phi_for_loop -------------------------------------
2728 // Insert phi(lp_entry_val, back_edge_val) at use->in(idx) for loop lp if phi does not already exist
2729 void PhaseIdealLoop::insert_phi_for_loop( Node* use, uint idx, Node* lp_entry_val, Node* back_edge_val, LoopNode* lp ) {
2730   Node *phi = PhiNode::make(lp, back_edge_val);
2731   phi->set_req(LoopNode::EntryControl, lp_entry_val);
2732   // Use existing phi if it already exists
2733   Node *hit = _igvn.hash_find_insert(phi);
2734   if( hit == NULL ) {
2735     _igvn.register_new_node_with_optimizer(phi);
2736     set_ctrl(phi, lp);
2737   } else {
2738     // Remove the new phi from the graph and use the hit
2739     _igvn.remove_dead_node(phi);
2740     phi = hit;
2741   }
2742   _igvn.replace_input_of(use, idx, phi);
2743 }
2744 
2745 #ifdef ASSERT
2746 //------------------------------ is_valid_loop_partition -------------------------------------
2747 // Validate the loop partition sets: peel and not_peel
2748 bool PhaseIdealLoop::is_valid_loop_partition( IdealLoopTree *loop, VectorSet& peel, Node_List& peel_list,
2749                                               VectorSet& not_peel ) {
2750   uint i;
2751   // Check that peel_list entries are in the peel set
2752   for (i = 0; i < peel_list.size(); i++) {
2753     if (!peel.test(peel_list.at(i)->_idx)) {
2754       return false;
2755     }
2756   }
2757   // Check at loop members are in one of peel set or not_peel set
2758   for (i = 0; i < loop->_body.size(); i++ ) {
2759     Node *def  = loop->_body.at(i);
2760     uint di = def->_idx;
2761     // Check that peel set elements are in peel_list
2762     if (peel.test(di)) {
2763       if (not_peel.test(di)) {
2764         return false;
2765       }
2766       // Must be in peel_list also
2767       bool found = false;
2768       for (uint j = 0; j < peel_list.size(); j++) {
2769         if (peel_list.at(j)->_idx == di) {
2770           found = true;
2771           break;
2772         }
2773       }
2774       if (!found) {
2775         return false;
2776       }
2777     } else if (not_peel.test(di)) {
2778       if (peel.test(di)) {
2779         return false;
2780       }
2781     } else {
2782       return false;
2783     }
2784   }
2785   return true;
2786 }
2787 
2788 //------------------------------ is_valid_clone_loop_exit_use -------------------------------------
2789 // Ensure a use outside of loop is of the right form
2790 bool PhaseIdealLoop::is_valid_clone_loop_exit_use( IdealLoopTree *loop, Node* use, uint exit_idx) {
2791   Node *use_c = has_ctrl(use) ? get_ctrl(use) : use;
2792   return (use->is_Phi() &&
2793           use_c->is_Region() && use_c->req() == 3 &&
2794           (use_c->in(exit_idx)->Opcode() == Op_IfTrue ||
2795            use_c->in(exit_idx)->Opcode() == Op_IfFalse ||
2796            use_c->in(exit_idx)->Opcode() == Op_JumpProj) &&
2797           loop->is_member( get_loop( use_c->in(exit_idx)->in(0) ) ) );
2798 }
2799 
2800 //------------------------------ is_valid_clone_loop_form -------------------------------------
2801 // Ensure that all uses outside of loop are of the right form
2802 bool PhaseIdealLoop::is_valid_clone_loop_form( IdealLoopTree *loop, Node_List& peel_list,
2803                                                uint orig_exit_idx, uint clone_exit_idx) {
2804   uint len = peel_list.size();
2805   for (uint i = 0; i < len; i++) {
2806     Node *def = peel_list.at(i);
2807 
2808     for (DUIterator_Fast jmax, j = def->fast_outs(jmax); j < jmax; j++) {
2809       Node *use = def->fast_out(j);
2810       Node *use_c = has_ctrl(use) ? get_ctrl(use) : use;
2811       if (!loop->is_member(get_loop(use_c))) {
2812         // use is not in the loop, check for correct structure
2813         if (use->in(0) == def) {
2814           // Okay
2815         } else if (!is_valid_clone_loop_exit_use(loop, use, orig_exit_idx)) {
2816           return false;
2817         }
2818       }
2819     }
2820   }
2821   return true;
2822 }
2823 #endif
2824 
2825 //------------------------------ partial_peel -------------------------------------
2826 // Partially peel (aka loop rotation) the top portion of a loop (called
2827 // the peel section below) by cloning it and placing one copy just before
2828 // the new loop head and the other copy at the bottom of the new loop.
2829 //
2830 //    before                       after                where it came from
2831 //
2832 //    stmt1                        stmt1
2833 //  loop:                          stmt2                     clone
2834 //    stmt2                        if condA goto exitA       clone
2835 //    if condA goto exitA        new_loop:                   new
2836 //    stmt3                        stmt3                     clone
2837 //    if !condB goto loop          if condB goto exitB       clone
2838 //  exitB:                         stmt2                     orig
2839 //    stmt4                        if !condA goto new_loop   orig
2840 //  exitA:                         goto exitA
2841 //                               exitB:
2842 //                                 stmt4
2843 //                               exitA:
2844 //
2845 // Step 1: find the cut point: an exit test on probable
2846 //         induction variable.
2847 // Step 2: schedule (with cloning) operations in the peel
2848 //         section that can be executed after the cut into
2849 //         the section that is not peeled.  This may need
2850 //         to clone operations into exit blocks.  For
2851 //         instance, a reference to A[i] in the not-peel
2852 //         section and a reference to B[i] in an exit block
2853 //         may cause a left-shift of i by 2 to be placed
2854 //         in the peel block.  This step will clone the left
2855 //         shift into the exit block and sink the left shift
2856 //         from the peel to the not-peel section.
2857 // Step 3: clone the loop, retarget the control, and insert
2858 //         phis for values that are live across the new loop
2859 //         head.  This is very dependent on the graph structure
2860 //         from clone_loop.  It creates region nodes for
2861 //         exit control and associated phi nodes for values
2862 //         flow out of the loop through that exit.  The region
2863 //         node is dominated by the clone's control projection.
2864 //         So the clone's peel section is placed before the
2865 //         new loop head, and the clone's not-peel section is
2866 //         forms the top part of the new loop.  The original
2867 //         peel section forms the tail of the new loop.
2868 // Step 4: update the dominator tree and recompute the
2869 //         dominator depth.
2870 //
2871 //                   orig
2872 //
2873 //                   stmt1
2874 //                     |
2875 //                     v
2876 //               loop predicate
2877 //                     |
2878 //                     v
2879 //                   loop<----+
2880 //                     |      |
2881 //                   stmt2    |
2882 //                     |      |
2883 //                     v      |
2884 //                    ifA     |
2885 //                   / |      |
2886 //                  v  v      |
2887 //               false true   ^  <-- last_peel
2888 //               /     |      |
2889 //              /   ===|==cut |
2890 //             /     stmt3    |  <-- first_not_peel
2891 //            /        |      |
2892 //            |        v      |
2893 //            v       ifB     |
2894 //          exitA:   / \      |
2895 //                  /   \     |
2896 //                 v     v    |
2897 //               false true   |
2898 //               /       \    |
2899 //              /         ----+
2900 //             |
2901 //             v
2902 //           exitB:
2903 //           stmt4
2904 //
2905 //
2906 //            after clone loop
2907 //
2908 //                   stmt1
2909 //                     |
2910 //                     v
2911 //               loop predicate
2912 //                 /       \
2913 //        clone   /         \   orig
2914 //               /           \
2915 //              /             \
2916 //             v               v
2917 //   +---->loop                loop<----+
2918 //   |      |                    |      |
2919 //   |    stmt2                stmt2    |
2920 //   |      |                    |      |
2921 //   |      v                    v      |
2922 //   |      ifA                 ifA     |
2923 //   |      | \                / |      |
2924 //   |      v  v              v  v      |
2925 //   ^    true  false      false true   ^  <-- last_peel
2926 //   |      |   ^   \       /    |      |
2927 //   | cut==|==  \   \     /  ===|==cut |
2928 //   |    stmt3   \   \   /    stmt3    |  <-- first_not_peel
2929 //   |      |    dom   | |       |      |
2930 //   |      v      \  1v v2      v      |
2931 //   |      ifB     regionA     ifB     |
2932 //   |      / \        |       / \      |
2933 //   |     /   \       v      /   \     |
2934 //   |    v     v    exitA:  v     v    |
2935 //   |    true  false      false true   |
2936 //   |    /     ^   \      /       \    |
2937 //   +----       \   \    /         ----+
2938 //               dom  \  /
2939 //                 \  1v v2
2940 //                  regionB
2941 //                     |
2942 //                     v
2943 //                   exitB:
2944 //                   stmt4
2945 //
2946 //
2947 //           after partial peel
2948 //
2949 //                  stmt1
2950 //                     |
2951 //                     v
2952 //               loop predicate
2953 //                 /
2954 //        clone   /             orig
2955 //               /          TOP
2956 //              /             \
2957 //             v               v
2958 //    TOP->loop                loop----+
2959 //          |                    |      |
2960 //        stmt2                stmt2    |
2961 //          |                    |      |
2962 //          v                    v      |
2963 //          ifA                 ifA     |
2964 //          | \                / |      |
2965 //          v  v              v  v      |
2966 //        true  false      false true   |     <-- last_peel
2967 //          |   ^   \       /    +------|---+
2968 //  +->newloop   \   \     /  === ==cut |   |
2969 //  |     stmt3   \   \   /     TOP     |   |
2970 //  |       |    dom   | |      stmt3   |   | <-- first_not_peel
2971 //  |       v      \  1v v2      v      |   |
2972 //  |       ifB     regionA     ifB     ^   v
2973 //  |       / \        |       / \      |   |
2974 //  |      /   \       v      /   \     |   |
2975 //  |     v     v    exitA:  v     v    |   |
2976 //  |     true  false      false true   |   |
2977 //  |     /     ^   \      /       \    |   |
2978 //  |    |       \   \    /         v   |   |
2979 //  |    |       dom  \  /         TOP  |   |
2980 //  |    |         \  1v v2             |   |
2981 //  ^    v          regionB             |   |
2982 //  |    |             |                |   |
2983 //  |    |             v                ^   v
2984 //  |    |           exitB:             |   |
2985 //  |    |           stmt4              |   |
2986 //  |    +------------>-----------------+   |
2987 //  |                                       |
2988 //  +-----------------<---------------------+
2989 //
2990 //
2991 //              final graph
2992 //
2993 //                  stmt1
2994 //                    |
2995 //                    v
2996 //               loop predicate
2997 //                    |
2998 //                    v
2999 //                  stmt2 clone
3000 //                    |
3001 //                    v
3002 //         ........> ifA clone
3003 //         :        / |
3004 //        dom      /  |
3005 //         :      v   v
3006 //         :  false   true
3007 //         :  |       |
3008 //         :  |       v
3009 //         :  |    newloop<-----+
3010 //         :  |        |        |
3011 //         :  |     stmt3 clone |
3012 //         :  |        |        |
3013 //         :  |        v        |
3014 //         :  |       ifB       |
3015 //         :  |      / \        |
3016 //         :  |     v   v       |
3017 //         :  |  false true     |
3018 //         :  |   |     |       |
3019 //         :  |   v    stmt2    |
3020 //         :  | exitB:  |       |
3021 //         :  | stmt4   v       |
3022 //         :  |       ifA orig  |
3023 //         :  |      /  \       |
3024 //         :  |     /    \      |
3025 //         :  |    v     v      |
3026 //         :  |  false  true    |
3027 //         :  |  /        \     |
3028 //         :  v  v         -----+
3029 //          RegionA
3030 //             |
3031 //             v
3032 //           exitA
3033 //
3034 bool PhaseIdealLoop::partial_peel( IdealLoopTree *loop, Node_List &old_new ) {
3035 
3036   assert(!loop->_head->is_CountedLoop(), "Non-counted loop only");
3037   if (!loop->_head->is_Loop()) {
3038     return false;
3039   }
3040   LoopNode *head = loop->_head->as_Loop();
3041 
3042   if (head->is_partial_peel_loop() || head->partial_peel_has_failed()) {
3043     return false;
3044   }
3045 
3046   // Check for complex exit control
3047   for (uint ii = 0; ii < loop->_body.size(); ii++) {
3048     Node *n = loop->_body.at(ii);
3049     int opc = n->Opcode();
3050     if (n->is_Call()        ||
3051         opc == Op_Catch     ||
3052         opc == Op_CatchProj ||
3053         opc == Op_Jump      ||
3054         opc == Op_JumpProj) {
3055 #ifndef PRODUCT
3056       if (TracePartialPeeling) {
3057         tty->print_cr("\nExit control too complex: lp: %d", head->_idx);
3058       }
3059 #endif
3060       return false;
3061     }
3062   }
3063 
3064   int dd = dom_depth(head);
3065 
3066   // Step 1: find cut point
3067 
3068   // Walk up dominators to loop head looking for first loop exit
3069   // which is executed on every path thru loop.
3070   IfNode *peel_if = NULL;
3071   IfNode *peel_if_cmpu = NULL;
3072 
3073   Node *iff = loop->tail();
3074   while (iff != head) {
3075     if (iff->is_If()) {
3076       Node *ctrl = get_ctrl(iff->in(1));
3077       if (ctrl->is_top()) return false; // Dead test on live IF.
3078       // If loop-varying exit-test, check for induction variable
3079       if (loop->is_member(get_loop(ctrl)) &&
3080           loop->is_loop_exit(iff) &&
3081           is_possible_iv_test(iff)) {
3082         Node* cmp = iff->in(1)->in(1);
3083         if (cmp->Opcode() == Op_CmpI) {
3084           peel_if = iff->as_If();
3085         } else {
3086           assert(cmp->Opcode() == Op_CmpU, "must be CmpI or CmpU");
3087           peel_if_cmpu = iff->as_If();
3088         }
3089       }
3090     }
3091     iff = idom(iff);
3092   }
3093 
3094   // Prefer signed compare over unsigned compare.
3095   IfNode* new_peel_if = NULL;
3096   if (peel_if == NULL) {
3097     if (!PartialPeelAtUnsignedTests || peel_if_cmpu == NULL) {
3098       return false;   // No peel point found
3099     }
3100     new_peel_if = insert_cmpi_loop_exit(peel_if_cmpu, loop);
3101     if (new_peel_if == NULL) {
3102       return false;   // No peel point found
3103     }
3104     peel_if = new_peel_if;
3105   }
3106   Node* last_peel        = stay_in_loop(peel_if, loop);
3107   Node* first_not_peeled = stay_in_loop(last_peel, loop);
3108   if (first_not_peeled == NULL || first_not_peeled == head) {
3109     return false;
3110   }
3111 
3112 #ifndef PRODUCT
3113   if (TraceLoopOpts) {
3114     tty->print("PartialPeel  ");
3115     loop->dump_head();
3116   }
3117 
3118   if (TracePartialPeeling) {
3119     tty->print_cr("before partial peel one iteration");
3120     Node_List wl;
3121     Node* t = head->in(2);
3122     while (true) {
3123       wl.push(t);
3124       if (t == head) break;
3125       t = idom(t);
3126     }
3127     while (wl.size() > 0) {
3128       Node* tt = wl.pop();
3129       tt->dump();
3130       if (tt == last_peel) tty->print_cr("-- cut --");
3131     }
3132   }
3133 #endif
3134   ResourceArea *area = Thread::current()->resource_area();
3135   VectorSet peel(area);
3136   VectorSet not_peel(area);
3137   Node_List peel_list(area);
3138   Node_List worklist(area);
3139   Node_List sink_list(area);
3140 
3141   if (!may_require_nodes(loop->est_loop_clone_sz(2))) {
3142     return false;
3143   }
3144 
3145   // Set of cfg nodes to peel are those that are executable from
3146   // the head through last_peel.
3147   assert(worklist.size() == 0, "should be empty");
3148   worklist.push(head);
3149   peel.set(head->_idx);
3150   while (worklist.size() > 0) {
3151     Node *n = worklist.pop();
3152     if (n != last_peel) {
3153       for (DUIterator_Fast jmax, j = n->fast_outs(jmax); j < jmax; j++) {
3154         Node* use = n->fast_out(j);
3155         if (use->is_CFG() &&
3156             loop->is_member(get_loop(use)) &&
3157             !peel.test_set(use->_idx)) {
3158           worklist.push(use);
3159         }
3160       }
3161     }
3162   }
3163 
3164   // Set of non-cfg nodes to peel are those that are control
3165   // dependent on the cfg nodes.
3166   uint i;
3167   for(i = 0; i < loop->_body.size(); i++ ) {
3168     Node *n = loop->_body.at(i);
3169     Node *n_c = has_ctrl(n) ? get_ctrl(n) : n;
3170     if (peel.test(n_c->_idx)) {
3171       peel.set(n->_idx);
3172     } else {
3173       not_peel.set(n->_idx);
3174     }
3175   }
3176 
3177   // Step 2: move operations from the peeled section down into the
3178   //         not-peeled section
3179 
3180   // Get a post order schedule of nodes in the peel region
3181   // Result in right-most operand.
3182   scheduled_nodelist(loop, peel, peel_list );
3183 
3184   assert(is_valid_loop_partition(loop, peel, peel_list, not_peel), "bad partition");
3185 
3186   // For future check for too many new phis
3187   uint old_phi_cnt = 0;
3188   for (DUIterator_Fast jmax, j = head->fast_outs(jmax); j < jmax; j++) {
3189     Node* use = head->fast_out(j);
3190     if (use->is_Phi()) old_phi_cnt++;
3191   }
3192 
3193 #ifndef PRODUCT
3194   if (TracePartialPeeling) {
3195     tty->print_cr("\npeeled list");
3196   }
3197 #endif
3198 
3199   // Evacuate nodes in peel region into the not_peeled region if possible
3200   uint new_phi_cnt = 0;
3201   uint cloned_for_outside_use = 0;
3202   for (i = 0; i < peel_list.size();) {
3203     Node* n = peel_list.at(i);
3204 #ifndef PRODUCT
3205   if (TracePartialPeeling) n->dump();
3206 #endif
3207     bool incr = true;
3208     if ( !n->is_CFG() ) {
3209 
3210       if ( has_use_in_set(n, not_peel) ) {
3211 
3212         // If not used internal to the peeled region,
3213         // move "n" from peeled to not_peeled region.
3214 
3215         if ( !has_use_internal_to_set(n, peel, loop) ) {
3216 
3217           // if not pinned and not a load (which maybe anti-dependent on a store)
3218           // and not a CMove (Matcher expects only bool->cmove).
3219           if (n->in(0) == NULL && !n->is_Load() && !n->is_CMove()) {
3220             cloned_for_outside_use += clone_for_use_outside_loop( loop, n, worklist );
3221             sink_list.push(n);
3222             peel     >>= n->_idx; // delete n from peel set.
3223             not_peel <<= n->_idx; // add n to not_peel set.
3224             peel_list.remove(i);
3225             incr = false;
3226 #ifndef PRODUCT
3227             if (TracePartialPeeling) {
3228               tty->print_cr("sink to not_peeled region: %d newbb: %d",
3229                             n->_idx, get_ctrl(n)->_idx);
3230             }
3231 #endif
3232           }
3233         } else {
3234           // Otherwise check for special def-use cases that span
3235           // the peel/not_peel boundary such as bool->if
3236           clone_for_special_use_inside_loop( loop, n, not_peel, sink_list, worklist );
3237           new_phi_cnt++;
3238         }
3239       }
3240     }
3241     if (incr) i++;
3242   }
3243 
3244   if (new_phi_cnt > old_phi_cnt + PartialPeelNewPhiDelta) {
3245 #ifndef PRODUCT
3246     if (TracePartialPeeling) {
3247       tty->print_cr("\nToo many new phis: %d  old %d new cmpi: %c",
3248                     new_phi_cnt, old_phi_cnt, new_peel_if != NULL?'T':'F');
3249     }
3250 #endif
3251     if (new_peel_if != NULL) {
3252       remove_cmpi_loop_exit(new_peel_if, loop);
3253     }
3254     // Inhibit more partial peeling on this loop
3255     assert(!head->is_partial_peel_loop(), "not partial peeled");
3256     head->mark_partial_peel_failed();
3257     if (cloned_for_outside_use > 0) {
3258       // Terminate this round of loop opts because
3259       // the graph outside this loop was changed.
3260       C->set_major_progress();
3261       return true;
3262     }
3263     return false;
3264   }
3265 
3266   // Step 3: clone loop, retarget control, and insert new phis
3267 
3268   // Create new loop head for new phis and to hang
3269   // the nodes being moved (sinked) from the peel region.
3270   LoopNode* new_head = new LoopNode(last_peel, last_peel);
3271   new_head->set_unswitch_count(head->unswitch_count()); // Preserve
3272   _igvn.register_new_node_with_optimizer(new_head);
3273   assert(first_not_peeled->in(0) == last_peel, "last_peel <- first_not_peeled");
3274   _igvn.replace_input_of(first_not_peeled, 0, new_head);
3275   set_loop(new_head, loop);
3276   loop->_body.push(new_head);
3277   not_peel.set(new_head->_idx);
3278   set_idom(new_head, last_peel, dom_depth(first_not_peeled));
3279   set_idom(first_not_peeled, new_head, dom_depth(first_not_peeled));
3280 
3281   while (sink_list.size() > 0) {
3282     Node* n = sink_list.pop();
3283     set_ctrl(n, new_head);
3284   }
3285 
3286   assert(is_valid_loop_partition(loop, peel, peel_list, not_peel), "bad partition");
3287 
3288   clone_loop(loop, old_new, dd, IgnoreStripMined);
3289 
3290   const uint clone_exit_idx = 1;
3291   const uint orig_exit_idx  = 2;
3292   assert(is_valid_clone_loop_form( loop, peel_list, orig_exit_idx, clone_exit_idx ), "bad clone loop");
3293 
3294   Node* head_clone             = old_new[head->_idx];
3295   LoopNode* new_head_clone     = old_new[new_head->_idx]->as_Loop();
3296   Node* orig_tail_clone        = head_clone->in(2);
3297 
3298   // Add phi if "def" node is in peel set and "use" is not
3299 
3300   for(i = 0; i < peel_list.size(); i++ ) {
3301     Node *def  = peel_list.at(i);
3302     if (!def->is_CFG()) {
3303       for (DUIterator_Fast jmax, j = def->fast_outs(jmax); j < jmax; j++) {
3304         Node *use = def->fast_out(j);
3305         if (has_node(use) && use->in(0) != C->top() &&
3306             (!peel.test(use->_idx) ||
3307              (use->is_Phi() && use->in(0) == head)) ) {
3308           worklist.push(use);
3309         }
3310       }
3311       while( worklist.size() ) {
3312         Node *use = worklist.pop();
3313         for (uint j = 1; j < use->req(); j++) {
3314           Node* n = use->in(j);
3315           if (n == def) {
3316 
3317             // "def" is in peel set, "use" is not in peel set
3318             // or "use" is in the entry boundary (a phi) of the peel set
3319 
3320             Node* use_c = has_ctrl(use) ? get_ctrl(use) : use;
3321 
3322             if ( loop->is_member(get_loop( use_c )) ) {
3323               // use is in loop
3324               if (old_new[use->_idx] != NULL) { // null for dead code
3325                 Node* use_clone = old_new[use->_idx];
3326                 _igvn.replace_input_of(use, j, C->top());
3327                 insert_phi_for_loop( use_clone, j, old_new[def->_idx], def, new_head_clone );
3328               }
3329             } else {
3330               assert(is_valid_clone_loop_exit_use(loop, use, orig_exit_idx), "clone loop format");
3331               // use is not in the loop, check if the live range includes the cut
3332               Node* lp_if = use_c->in(orig_exit_idx)->in(0);
3333               if (not_peel.test(lp_if->_idx)) {
3334                 assert(j == orig_exit_idx, "use from original loop");
3335                 insert_phi_for_loop( use, clone_exit_idx, old_new[def->_idx], def, new_head_clone );
3336               }
3337             }
3338           }
3339         }
3340       }
3341     }
3342   }
3343 
3344   // Step 3b: retarget control
3345 
3346   // Redirect control to the new loop head if a cloned node in
3347   // the not_peeled region has control that points into the peeled region.
3348   // This necessary because the cloned peeled region will be outside
3349   // the loop.
3350   //                            from    to
3351   //          cloned-peeled    <---+
3352   //    new_head_clone:            |    <--+
3353   //          cloned-not_peeled  in(0)    in(0)
3354   //          orig-peeled
3355 
3356   for(i = 0; i < loop->_body.size(); i++ ) {
3357     Node *n = loop->_body.at(i);
3358     if (!n->is_CFG()           && n->in(0) != NULL        &&
3359         not_peel.test(n->_idx) && peel.test(n->in(0)->_idx)) {
3360       Node* n_clone = old_new[n->_idx];
3361       _igvn.replace_input_of(n_clone, 0, new_head_clone);
3362     }
3363   }
3364 
3365   // Backedge of the surviving new_head (the clone) is original last_peel
3366   _igvn.replace_input_of(new_head_clone, LoopNode::LoopBackControl, last_peel);
3367 
3368   // Cut first node in original not_peel set
3369   _igvn.rehash_node_delayed(new_head);                     // Multiple edge updates:
3370   new_head->set_req(LoopNode::EntryControl,    C->top());  //   use rehash_node_delayed / set_req instead of
3371   new_head->set_req(LoopNode::LoopBackControl, C->top());  //   multiple replace_input_of calls
3372 
3373   // Copy head_clone back-branch info to original head
3374   // and remove original head's loop entry and
3375   // clone head's back-branch
3376   _igvn.rehash_node_delayed(head); // Multiple edge updates
3377   head->set_req(LoopNode::EntryControl,    head_clone->in(LoopNode::LoopBackControl));
3378   head->set_req(LoopNode::LoopBackControl, C->top());
3379   _igvn.replace_input_of(head_clone, LoopNode::LoopBackControl, C->top());
3380 
3381   // Similarly modify the phis
3382   for (DUIterator_Fast kmax, k = head->fast_outs(kmax); k < kmax; k++) {
3383     Node* use = head->fast_out(k);
3384     if (use->is_Phi() && use->outcnt() > 0) {
3385       Node* use_clone = old_new[use->_idx];
3386       _igvn.rehash_node_delayed(use); // Multiple edge updates
3387       use->set_req(LoopNode::EntryControl,    use_clone->in(LoopNode::LoopBackControl));
3388       use->set_req(LoopNode::LoopBackControl, C->top());
3389       _igvn.replace_input_of(use_clone, LoopNode::LoopBackControl, C->top());
3390     }
3391   }
3392 
3393   // Step 4: update dominator tree and dominator depth
3394 
3395   set_idom(head, orig_tail_clone, dd);
3396   recompute_dom_depth();
3397 
3398   // Inhibit more partial peeling on this loop
3399   new_head_clone->set_partial_peel_loop();
3400   C->set_major_progress();
3401   loop->record_for_igvn();
3402 
3403 #ifndef PRODUCT
3404   if (TracePartialPeeling) {
3405     tty->print_cr("\nafter partial peel one iteration");
3406     Node_List wl(area);
3407     Node* t = last_peel;
3408     while (true) {
3409       wl.push(t);
3410       if (t == head_clone) break;
3411       t = idom(t);
3412     }
3413     while (wl.size() > 0) {
3414       Node* tt = wl.pop();
3415       if (tt == head) tty->print_cr("orig head");
3416       else if (tt == new_head_clone) tty->print_cr("new head");
3417       else if (tt == head_clone) tty->print_cr("clone head");
3418       tt->dump();
3419     }
3420   }
3421 #endif
3422   return true;
3423 }
3424 
3425 //------------------------------reorg_offsets----------------------------------
3426 // Reorganize offset computations to lower register pressure.  Mostly
3427 // prevent loop-fallout uses of the pre-incremented trip counter (which are
3428 // then alive with the post-incremented trip counter forcing an extra
3429 // register move)
3430 void PhaseIdealLoop::reorg_offsets(IdealLoopTree *loop) {
3431   // Perform it only for canonical counted loops.
3432   // Loop's shape could be messed up by iteration_split_impl.
3433   if (!loop->_head->is_CountedLoop())
3434     return;
3435   if (!loop->_head->as_Loop()->is_valid_counted_loop())
3436     return;
3437 
3438   CountedLoopNode *cl = loop->_head->as_CountedLoop();
3439   CountedLoopEndNode *cle = cl->loopexit();
3440   Node *exit = cle->proj_out(false);
3441   Node *phi = cl->phi();
3442 
3443   // Check for the special case when using the pre-incremented trip-counter on
3444   // the fall-out  path (forces the pre-incremented  and post-incremented trip
3445   // counter to be live  at the same time).  Fix this by  adjusting to use the
3446   // post-increment trip counter.
3447 
3448   bool progress = true;
3449   while (progress) {
3450     progress = false;
3451     for (DUIterator_Fast imax, i = phi->fast_outs(imax); i < imax; i++) {
3452       Node* use = phi->fast_out(i);   // User of trip-counter
3453       if (!has_ctrl(use))  continue;
3454       Node *u_ctrl = get_ctrl(use);
3455       if (use->is_Phi()) {
3456         u_ctrl = NULL;
3457         for (uint j = 1; j < use->req(); j++)
3458           if (use->in(j) == phi)
3459             u_ctrl = dom_lca(u_ctrl, use->in(0)->in(j));
3460       }
3461       IdealLoopTree *u_loop = get_loop(u_ctrl);
3462       // Look for loop-invariant use
3463       if (u_loop == loop) continue;
3464       if (loop->is_member(u_loop)) continue;
3465       // Check that use is live out the bottom.  Assuming the trip-counter
3466       // update is right at the bottom, uses of of the loop middle are ok.
3467       if (dom_lca(exit, u_ctrl) != exit) continue;
3468       // Hit!  Refactor use to use the post-incremented tripcounter.
3469       // Compute a post-increment tripcounter.
3470       Node* c = exit;
3471       if (cl->is_strip_mined()) {
3472         IdealLoopTree* outer_loop = get_loop(cl->outer_loop());
3473         if (!outer_loop->is_member(u_loop)) {
3474           c = cl->outer_loop_exit();
3475         }
3476       }
3477       Node *opaq = new Opaque2Node(C, cle->incr());
3478       register_new_node(opaq, c);
3479       Node *neg_stride = _igvn.intcon(-cle->stride_con());
3480       set_ctrl(neg_stride, C->root());
3481       Node *post = new AddINode(opaq, neg_stride);
3482       register_new_node(post, c);
3483       _igvn.rehash_node_delayed(use);
3484       for (uint j = 1; j < use->req(); j++) {
3485         if (use->in(j) == phi)
3486           use->set_req(j, post);
3487       }
3488       // Since DU info changed, rerun loop
3489       progress = true;
3490       break;
3491     }
3492   }
3493 
3494 }