RFR: 8358679: [asan] vmTestbase/nsk/jvmti tests show memory issues

[Chris Plummer]

On Tue, 8 Jul 2025 00:01:50 GMT, Alex Menkov <amenkov at openjdk.org> wrote:

> The fix updates several tests ASAN complains about.
> Tests for RawMonitors intentionally passes bad pointer to tested function, they should not be run with ASAN.
> Test vmTestbase/nsk/jvmti/scenarios/events/EM07/em07t002 verifies COMPILED_METHOD_LOAD/COMPILED_METHOD_UNLOAD event and uses RawMonitor for synchronization. It destroys the monitor, but some events may arrive later (before VM exits). Usually JVMTI tests do not destroy synchronization monitors, removed destruction here.

test/hotspot/jtreg/vmTestbase/nsk/jvmti/RawMonitorEnter/rawmonenter003/TestDescription.java line 41:

> 39:  *
> 40:  * @comment the test intentionally passes bad argument to the function to verify the functionality,
> 41:             it causes false positive from ASAN lib

Suggestion:

 * @comment The test intentionally passes a bad argument to the function to verify error checking,
            which causes a false positive from the ASAN lib


Same comment for the other tests.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/26176#discussion_r2191478686