JDK-8346954 : [JMH] jdk.incubator.vector.MaskedLogicOpts fails due to IndexOutOfBoundsException
Type:Bug
Component:hotspot
Sub-Component:compiler
Affected Version:21,24,25
Priority:P4
Status:Resolved
Resolution:Fixed
CPU:generic
Submitted:2025-01-03
Updated:2025-03-20
Resolved:2025-03-06
The Version table provides details related to the release that this issue/RFE will be addressed.
Unresolved : Release in which this issue/RFE will be addressed. Resolved: Release in which this issue/RFE has been resolved. Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.
Suite MaskedLogicOpts.maskedLogicOperationsLong512() crashed on both x86 and AArch64 with the following error:
```
java.lang.IndexOutOfBoundsException: Index 252 out of bounds for length 249
```
A pull request was submitted for review.
Branch: master
URL: https://git.openjdk.org/jdk/pull/22963
Date: 2025-01-08 09:04:47 +0000
05-02-2025
Here is a PR from my colleague to fix the test: https://github.com/openjdk/jdk/pull/22963
14-01-2025
Paging [~jbhateja] who's the author of this test.
13-01-2025
ILW = Microbenchmark fails with IOOBE (test bug), single microbenchmark, no known workaround = MLH = P4
13-01-2025
Thanks [~xgong]. I had a quick look and it's indeed easy reproduce and a test bug. The problem is that LongVector.fromArray throws an IndexOutOfBoundsException if offset+N < 0 or offset+N >= a.length for any lane N in the vector:
https://github.com/openjdk/jdk/blob/82e2a791225a289ba32360bf415274c4b48b9e00/test/micro/org/openjdk/bench/jdk/incubator/vector/MaskedLogicOpts.java#L309
https://docs.oracle.com/en/java/javase/17/docs/api/jdk.incubator.vector/jdk/incubator/vector/LongVector.html#fromArray(jdk.incubator.vector.VectorSpecies,long%5B%5D,int)
In this case, long256_arr_idx is 252 which out of bounds for length 249. The computation of long256_arr_idx (and probably others) needs to be fixed:
https://github.com/openjdk/jdk/blob/82e2a791225a289ba32360bf415274c4b48b9e00/test/micro/org/openjdk/bench/jdk/incubator/vector/MaskedLogicOpts.java#L132
13-01-2025
[~thartmann] Thanks for looking at this ticket.
// [~xgong] We don't see this failure in our testing. Could you please provide more
// information about the environment? VM flags, architecture, jtreg/jmh log file with
// stack trace etc. would be useful. Thanks!
This issue exists on all kind of architecture like AArch64 and X86. The architecture we tested on includes ARM SVE 128-bit vector length, and X86 AVX-512 system.
Following is the log details:
```
# JMH version: 1.37
# VM version: JDK 25-internal, OpenJDK 64-Bit Server VM, 25-internal-git-f1d85ab3e61
# VM invoker: jdk/bin/java
# VM options: --add-modules=jdk.incubator.vector -Xms8g -Xmx8g -XX:+AlwaysPreTouch -XX:ObjectAlignmentInBytes=64
# Blackhole mode: compiler (auto-detected, use -Djmh.blackhole.autoDetect=false to disable)
# Warmup: 1 iterations, 10 s each
# Measurement: 1 iterations, 10 s each
# Timeout: 10 min per iteration
# Threads: 1 thread, will synchronize iterations
# Benchmark mode: Throughput, ops/time
# Benchmark: org.openjdk.bench.jdk.incubator.vector.MaskedLogicOpts.maskedLogicOperationsLong512
# Parameters: (ARRAYLEN = 256)
# Run progress: 60.00% complete, ETA 00:06:10
# Fork: 1 of 1
WARNING: Using incubator modules: jdk.incubator.vector
WARNING: A terminally deprecated method in sun.misc.Unsafe has been called
WARNING: sun.misc.Unsafe::objectFieldOffset has been called by org.openjdk.jmh.util.Utils
WARNING: Please consider reporting this to the maintainers of class org.openjdk.jmh.util.Utils
WARNING: sun.misc.Unsafe::objectFieldOffset will be removed in a future release
# Warmup Iteration 1: <failure>
java.lang.IndexOutOfBoundsException: Index 252 out of bounds for length 249
at java.base/jdk.internal.util.Preconditions.outOfBounds(Preconditions.java:100)
at java.base/jdk.internal.util.Preconditions.outOfBoundsCheckIndex(Preconditions.java:106)
at java.base/jdk.internal.util.Preconditions.checkIndex(Preconditions.java:302)
at java.base/java.util.Objects.checkIndex(Objects.java:365)
at jdk.incubator.vector/jdk.incubator.vector.VectorIntrinsics.checkFromIndexSize(VectorIntrinsics.java:58)
at jdk.incubator.vector/jdk.incubator.vector.LongVector.fromArray(LongVector.java:2866)
at org.openjdk.bench.jdk.incubator.vector.MaskedLogicOpts.maskedLogicOperationsLongKernel(MaskedLogicOpts.java:246)
at org.openjdk.bench.jdk.incubator.vector.MaskedLogicOpts.maskedLogicOperationsLong512(MaskedLogicOpts.java:266)
at org.openjdk.bench.jdk.incubator.vector.jmh_generated.MaskedLogicOpts_maskedLogicOperationsLong512_jmhTest.maskedLogicOperationsLong512_thrpt_jmhStub(MaskedLogicOpts_maskedLogicOperationsLong512_jmhTest.java:125)
at org.openjdk.bench.jdk.incubator.vector.jmh_generated.MaskedLogicOpts_maskedLogicOperationsLong512_jmhTest.maskedLogicOperationsLong512_Throughput(MaskedLogicOpts_maskedLogicOperationsLong512_jmhTest.java:85)
at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
at java.base/java.lang.reflect.Method.invoke(Method.java:565)
at org.openjdk.jmh.runner.BenchmarkHandler$BenchmarkTask.call(BenchmarkHandler.java:527)
at org.openjdk.jmh.runner.BenchmarkHandler$BenchmarkTask.call(BenchmarkHandler.java:504)
at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:328)
at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:545)
at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:328)
at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1095)
at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:619)
at java.base/java.lang.Thread.run(Thread.java:1447)
```
Hope this could provide any help. Thanks!
Xiaohong
13-01-2025
[~xgong] We don't see this failure in our testing. Could you please provide more information about the environment? VM flags, architecture, jtreg/jmh log file with stack trace etc. would be useful. Thanks!
Update: Ah, it's a microbenchmark. We probably don't execute it very often.
13-01-2025
[~xgong]please update affects version indicating the JDK version impacted.
12-01-2025
The test bugs should filed against corresponding hotspot components.