RFR: 8321687: Test vmTestbase/nsk/jvmti/scenarios/contention/TC03/tc03t002/TestDescription.java failed: JVMTI_ERROR_THREAD_NOT_ALIVE

[Chris Plummer]

On Wed, 15 Oct 2025 20:38:14 GMT, Leonid Mesnik <lmesnik at openjdk.org> wrote:

> Test might fail with 
> 
> ----------System.out:(5/399)----------
> The following fake exception stacktrace is for failure analysis.
> nsk.share.Fake_Exception_for_RULE_Creation: (tc03t002.cpp:144) jvmti->GetCurrentContendedMonitor(threadList[pThread].thread, &monitor)
> at nsk_lvcomplain(nsk_tools.cpp:172)
> # ERROR: tc03t002.cpp, 144: jvmti->GetCurrentContendedMonitor(threadList[pThread].thread, &monitor)
> # jvmti error: code=15, name=JVMTI_ERROR_THREAD_NOT_ALIVE
> 
> if some of threads unexpectedly finishes during test execution. 
> 
> 
> It might happens only for some tests that are not started and verified by thread. So the fix is to skip them and verify only "Debugee" threads that might be in the deadlock.

Isn't this also an issue for tc03t001?

test/hotspot/jtreg/vmTestbase/nsk/jvmti/scenarios/contention/TC03/tc03t002/tc03t002.cpp line 48:

> 46: static int numberOfDeadlocks = 0;
> 47: 
> 48: static const char* thread_name_prefix = "Debugee Thread";

I think "Debugee Thread" here and in the java source need a comment indicating that the C and java code need to be kept in sync. The java code should probably also assign it to a final field.

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

PR Review: https://git.openjdk.org/jdk/pull/27831#pullrequestreview-3342512588
PR Review Comment: https://git.openjdk.org/jdk/pull/27831#discussion_r2434094541