JDK-8291649 : multiple tests failing with -Xcomp after JDK-8290034
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 20
  • Priority: P2
  • Status: Open
  • Resolution: Unresolved
  • CPU: x86_64
  • Submitted: 2022-08-01
  • Updated: 2022-08-21
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.

To download the current JDK release, click here.
JDK 20
20Unresolved
Related Reports
Relates :  
Relates :  
Sub Tasks
JDK-8291926 :  
JDK-8291990 :  
Description
java/lang/Long/BitTwiddle.java:

java.lang.RuntimeException: lzx: 1
	at BitTwiddle.main(BitTwiddle.java:107)
	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
	at java.base/java.lang.reflect.Method.invoke(Method.java:578)
	at com.sun.javatest.regtest.agent.MainActionHelper$AgentVMRunnable.run(MainActionHelper.java:312)
	at java.base/java.lang.Thread.run(Thread.java:1589)

java/util/zip/TestCRC32C.java:

java.lang.RuntimeException: Calculated checksum result was invalid. Expected e3069283, but got fa64cc41.
	at ChecksumBase.checkChecksum(ChecksumBase.java:92)
	at ChecksumBase.testBytes(ChecksumBase.java:56)
	at ChecksumBase.testAll(ChecksumBase.java:37)
	at TestCRC32C.main(TestCRC32C.java:37)
	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
	at java.base/java.lang.reflect.Method.invoke(Method.java:578)
	at com.sun.javatest.regtest.agent.MainActionHelper$AgentVMRunnable.run(MainActionHelper.java:312)
	at java.base/java.lang.Thread.run(Thread.java:1589)

java/util/zip/TestChecksum.java:
java.lang.RuntimeException: Calculated checksum result was invalid. Expected e3069283, but got fa64cc41.
	at ChecksumBase.checkChecksum(ChecksumBase.java:92)
	at ChecksumBase.testBytes(ChecksumBase.java:56)
	at ChecksumBase.testAll(ChecksumBase.java:37)
	at TestChecksum.main(TestChecksum.java:35)
	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
	at java.base/java.lang.reflect.Method.invoke(Method.java:578)
	at com.sun.javatest.regtest.agent.MainActionHelper$AgentVMRunnable.run(MainActionHelper.java:312)
	at java.base/java.lang.Thread.run(Thread.java:1589)

Comments
Hi all, I noticed it late, problem is in the constant value folding computation routine where bit reversal is not followed by byte reversal, will be posting a fix shortly.
21-08-2022

Problem-listed in jdk-20+10-542.
08-08-2022

The four tests were ProblemListed in jdk-20+10-522 using: JDK-8291926 ProblemList multiple tests in -Xcomp mode due to JDK-8291649 However, this latest failure sighting is in jdk-20+10-530-tier6. I don't see how that's possible. Update: I added the entries to the wrong ProblemList-Xcomp.txt file.
06-08-2022

Regressions normally start out at P2 so I'm bumping this bug from P3 -> P2. Also these failures are very noisy in the JDK20 CI. More than 300 sightings in just a few days.
04-08-2022

I think impact should be "high" because compiled code produces a wrong result. Raising priority to P3.
02-08-2022

ILW = multiple tests failing; with C2 -Xcomp only; -XX:DisableIntrinsic=_reverse_l -XX:DisableIntrinsic=_reverse_i = MLM = P4
02-08-2022

Possibly related to one of these changes: 8290034 Auto vectorize reverse bit operations. 8287794 Reverse*VNode::Identity problem [~jbhateja], please take a look.
01-08-2022