Skip to content

[GR-63358] AMD64ArrayIndexOfOp: Fix short jump across unknown alignment. #10916

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
graalvmbot merged 1 commit into from
Mar 25, 2025
Merged

[GR-63358] AMD64ArrayIndexOfOp: Fix short jump across unknown alignment. #10916

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 7 additions & 4 deletions compiler/src/jdk.graal.compiler/src/jdk/graal/compiler/lir/amd64/AMD64ArrayIndexOfOp.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@
*/
package jdk.graal.compiler.lir.amd64;

import static jdk.graal.compiler.asm.amd64.AMD64MacroAssembler.ExtendMode.ZERO_EXTEND;
import static jdk.vm.ci.amd64.AMD64.r8;
import static jdk.vm.ci.amd64.AMD64.rax;
import static jdk.vm.ci.amd64.AMD64.rcx;
Expand All @@ -33,7 +34,6 @@
import static jdk.vm.ci.code.ValueUtil.asRegister;
import static jdk.vm.ci.code.ValueUtil.isRegister;
import static jdk.vm.ci.code.ValueUtil.isStackSlot;
import static jdk.graal.compiler.asm.amd64.AMD64MacroAssembler.ExtendMode.ZERO_EXTEND;

import java.util.EnumSet;

Expand All @@ -56,7 +56,6 @@
import jdk.graal.compiler.lir.Opcode;
import jdk.graal.compiler.lir.asm.CompilationResultBuilder;
import jdk.graal.compiler.lir.gen.LIRGeneratorTool;

import jdk.vm.ci.amd64.AMD64.CPUFeature;
import jdk.vm.ci.amd64.AMD64Kind;
import jdk.vm.ci.code.Register;
Expand Down Expand Up @@ -431,8 +430,12 @@ public void emitCode(CompilationResultBuilder crb, AMD64MacroAssembler asm) {
asm.addq(index, bulkSize);

boolean bulkLoopShortJmp = !((variant == LIRGeneratorTool.ArrayIndexOfVariant.MatchRange && nValues == 4 || variant == LIRGeneratorTool.ArrayIndexOfVariant.Table) && stride.value > 1);
// check if there are enough array slots remaining for the bulk loop
asm.cmpqAndJcc(index, arrayLength, ConditionFlag.Greater, skipBulkVectorLoop, bulkLoopShortJmp);
/*
* Check if there are enough array slots remaining for the bulk loop. Note: The alignment
* following the cmpAndJcc can lead to a jump distance > 127. This prevents safely using a
* short jump.
*/
asm.cmpqAndJcc(index, arrayLength, ConditionFlag.Greater, skipBulkVectorLoop, false);

asm.align(preferredLoopAlignment(crb));
asm.bind(bulkVectorLoop);
Expand Down
Loading