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

[Chad Rakoczy]

On Thu, 11 Dec 2025 06:35:49 GMT, Leonid Mesnik <lmesnik at openjdk.org> wrote:

>> Chad Rakoczy has updated the pull request incrementally with two additional commits since the last revision:
>> 
>>  - Require load and unload events
>>  - Revert "Fix NMethodRelocationTest"
>>    
>>    This reverts commit dae91d8d12820c94c2c412f3e84935072e572595.
>
> test/hotspot/jtreg/serviceability/jvmti/NMethodRelocation/NMethodRelocationTest.java line 172:
> 
>> 170:         WHITE_BOX.lockCompilation();
>> 171: 
>> 172:         System.out.printf("Relocated nmethod from 0x%016x to 0x%016x%n", originalNMethod.code_begin, relocatedNMethod.code_begin);
> 
> I think it is better to wait until method_unload is generated rather then don't check it in the test.
> If tests times out while waiting for jvmti event we could analyze process dumps and logs trying to understand while jvmti event hasn't been posted. 
> it might be sense to call WB.fullGC(), check if jvmti event posted sleep some time and repeat in the loop.

Thanks for the suggestion. I posted a new revision that requires the events be published.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/28683#discussion_r2632547115