RFR: 8369150: NMethodRelocationTest fails when JVMTI events not published before JVM exit [v2]

Chad Rakoczy duke at openjdk.org
Mon Jan 5 21:15:53 UTC 2026


On Thu, 1 Jan 2026 19:54:31 GMT, Leonid Mesnik <lmesnik at openjdk.org> wrote:

>> I agree I don't think we should rely on stdout. I updated the test so all the event checking happens in the agent, and I kept some of the logging for debugging.
>
> Thank for moving logic into test process. I wonder if separate process is needed at all now?
> Might be it is possible to eliminate 
> `ProcessTools.createTestJavaProcessBuilder(`
> and just do ll checks in the test?

It required coordinating shutdown with the agent (since an exit code of zero still causes the test to fail) but I managed to remove the separate process.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/28683#discussion_r2662818496


More information about the serviceability-dev mailing list