[turbofan] Proper dead code elimination as regular reducer.

src/compiler/loop-peeling.cc

 // Copyright 2015 the V8 project authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "src/compiler/common-operator.h"
 #include "src/compiler/graph.h"
 #include "src/compiler/loop-peeling.h"
 #include "src/compiler/node.h"
 #include "src/compiler/node-marker.h"
 #include "src/compiler/node-properties.h"
@@ skipping 194 matching lines @@
   if (exits.size() != 1) return nullptr;  // not peelable currently.
 
   //============================================================================
   // Construct the peeled iteration.
   //============================================================================
   PeeledIterationImpl* iter = new (tmp_zone) PeeledIterationImpl(tmp_zone);
   size_t estimated_peeled_size =
       5 + (loop->TotalSize() + exits.size() + rets.size()) * 2;
   Peeling peeling(graph, tmp_zone, estimated_peeled_size, &iter->node_pairs_);
 
-  Node* dead = graph->NewNode(common->DeadControl());
+  Node* dead = graph->NewNode(common->Dead());
 
   // Map the loop header nodes to their entry values.
   for (Node* node : loop_tree->HeaderNodes(loop)) {
     peeling.Insert(node, node->InputAt(kAssumedLoopEntryIndex));
   }
 
   // Copy all the nodes of loop body for the peeled iteration.
   peeling.CopyNodes(graph, tmp_zone, dead, loop_tree->BodyNodes(loop));
 
   //============================================================================
@@ skipping 98 matching lines @@
       }
     }
   }
 
   return iter;
 }
 
 }  // namespace compiler
 }  // namespace internal
 }  // namespace v8