/*
 * Copyright (c) 2011, 2018, Oracle and/or its affiliates. All rights reserved.
 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
 *
 * This code is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 only, as
 * published by the Free Software Foundation.
 *
 * This code is distributed in the hope that it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
 * version 2 for more details (a copy is included in the LICENSE file that
 * accompanied this code).
 *
 * You should have received a copy of the GNU General Public License version
 * 2 along with this work; if not, write to the Free Software Foundation,
 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
 *
 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
 * or visit www.oracle.com if you need additional information or have any
 * questions.
 */


package org.graalvm.compiler.lir;

import org.graalvm.compiler.asm.Label;
import org.graalvm.compiler.core.common.cfg.AbstractBlockBase;
import org.graalvm.compiler.lir.StandardOp.BranchOp;
import org.graalvm.compiler.lir.StandardOp.JumpOp;

LIR instructions such as JumpOp and BranchOp need to reference their target AbstractBlockBase. However, direct references are not possible since the control flow graph (and therefore successors lists) can be changed by optimizations - and fixing the instructions is error prone. Therefore, we represent an edge to block B from block A via the tuple (A, successor-index-of-B). That is, indirectly by storing the index into the successor list of A. Note therefore that the successor list cannot be re-ordered.
/** * LIR instructions such as {@link JumpOp} and {@link BranchOp} need to reference their target * {@link AbstractBlockBase}. However, direct references are not possible since the control flow * graph (and therefore successors lists) can be changed by optimizations - and fixing the * instructions is error prone. Therefore, we represent an edge to block B from block A via the * tuple {@code (A, * successor-index-of-B)}. That is, indirectly by storing the index into the successor list of A. * Note therefore that the successor list cannot be re-ordered. */
public final class LabelRef { private final LIR lir; private final AbstractBlockBase<?> block; private final int suxIndex;
Returns a new reference to a successor of the given block.
Params:
  • block – The base block that contains the successor list.
  • suxIndex – The index of the successor.
Returns:The newly created label reference.
/** * Returns a new reference to a successor of the given block. * * @param block The base block that contains the successor list. * @param suxIndex The index of the successor. * @return The newly created label reference. */
public static LabelRef forSuccessor(final LIR lir, final AbstractBlockBase<?> block, final int suxIndex) { return new LabelRef(lir, block, suxIndex); }
Returns a new reference to a successor of the given block.
Params:
  • block – The base block that contains the successor list.
  • suxIndex – The index of the successor.
/** * Returns a new reference to a successor of the given block. * * @param block The base block that contains the successor list. * @param suxIndex The index of the successor. */
private LabelRef(final LIR lir, final AbstractBlockBase<?> block, final int suxIndex) { this.lir = lir; this.block = block; this.suxIndex = suxIndex; } public AbstractBlockBase<?> getSourceBlock() { return block; } public AbstractBlockBase<?> getTargetBlock() { return block.getSuccessors()[suxIndex]; } public Label label() { return ((StandardOp.LabelOp) lir.getLIRforBlock(getTargetBlock()).get(0)).getLabel(); } @Override public String toString() { return getSourceBlock() + " -> " + (suxIndex < block.getSuccessors().length ? getTargetBlock() : "?"); } }