RFR: 8309698: [S390X] Test hotspot/jtreg/runtime/NMT/VirtualAllocCommitMerge.java fails after JDK-8299089 [v2]

[Thomas Stuefe]

On Tue, 16 Jul 2024 02:01:29 GMT, Manjunath S Matti. <duke at openjdk.org> wrote:

>> On s390 the testcase "VirtualAllocCommitMerge.java" when run under Interpreter with option -Xint fails. But when run as compiled with option -Xcomp -Xbatch it Passes.
>
> Manjunath S Matti. has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains two additional commits since the last revision:
> 
>  - Merge branch 'openjdk:master' into NMT_s390_issue
>  - 8309698: runtime/NMT/VirtualAllocCommitMerge.java failed on s390x if interpreted.

I still disagree with the patch as it is now. The error should be analyzed, understood, and fixed. E.g. I don't get the relationship to JDK-8299089. What in JDK-8299089 can have caused stack walking on s390 to fail?

This fix is not terribly urgent, since the error just affects stacks in NMT virtual allocation printouts, and only some stacks at that. So, if you want to get the s390 CI clean, it would be absolutely fine to problem-list the test for s390. That is usually how we mute problems that we cannot fix immediately.

I just want to prevent uncommented "cargo-cult-programming" to slip into NMT tests: things we do but we don't know why. That does not help anyone, it just adds a layer of obfuscation.

Thanks, Thomas

-------------

Changes requested by stuefe (Reviewer).

PR Review: https://git.openjdk.org/jdk/pull/20096#pullrequestreview-2181926646