RFR: 8369150: NMethodRelocationTest fails when JVMTI events not published before JVM exit [v5]

[Leonid Mesnik]

On Mon, 5 Jan 2026 21:11:00 GMT, Chad Rakoczy <duke at openjdk.org> wrote:

>> [JDK-8369150](https://bugs.openjdk.org/browse/JDK-8369150)
>> 
>> The test checks for JVMTI `COMPILED_METHOD_LOAD` and `COMPILED_METHOD_UNLOAD` events to be published for a relocated nmethod. It would originally intermittently fail if the JVM exited before it had time to publish the events so now it loops and forces GCs to encourage event publishing. The test fails if the events are received in the incorrect order (such as an unload before a load) or if the correct events are not received and the test times out.
>
> Chad Rakoczy has updated the pull request incrementally with two additional commits since the last revision:
> 
>  - Use jvmti_common functions
>  - Remove seperate process

Thanks for update. The test looks simpler and work faster now. Please find my inline comments related ro AttachCurrentThread usage.

Here are some inline comments

test/hotspot/jtreg/serviceability/jvmti/NMethodRelocation/libNMethodRelocationTest.cpp line 39:

> 37:  * Helper function to update the shouldExit field in the Java test class.
> 38:  */
> 39: static void updateShouldExit() {

The simpler and more usual way is to have native variable
`static std::atomic<bool> should_exit;`
and poke it from hava code with 

 JNIEXPORT jboolean JNICALL
 java_..._shouldExitt(JNIEnv *env, jobject obj) {
    return should_exit.load();
 }

It is more compact and and simpler.
Also, I am not sure if AttachCurrentThread supposed to work for NonJavaThread or VM-intrenal thread, like compiler thread.

test/hotspot/jtreg/serviceability/jvmti/NMethodRelocation/libNMethodRelocationTest.cpp line 116:

> 114: 
> 115:     // Only track events for "compiledMethod"
> 116:     char* name = get_method_name(jvmti, env, method);

Can you please add new method
`get_method_name(jvmti, method);`
that uses 
`check_jvmti_error(jvmtiError err, const char* msg) `
and doesn't need JNI.

So no need to use `AttachCurrentThread`.

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

Changes requested by lmesnik (Reviewer).

PR Review: https://git.openjdk.org/jdk/pull/28683#pullrequestreview-3628832754
PR Review: https://git.openjdk.org/jdk/pull/28683#pullrequestreview-3628833420
PR Review Comment: https://git.openjdk.org/jdk/pull/28683#discussion_r2663164281
PR Review Comment: https://git.openjdk.org/jdk/pull/28683#discussion_r2663167608