On Tue, 29 Jun 2021 16:38:28 GMT, Leonid Mesnik <lmes...@openjdk.org> wrote:
>> The crash happens because nmethod might become a zombie before it is
>> enqueued in JvmtiDeferredEventQueue or after it is dequeued from it. The
>> crash is reproduced by serviceability/jvmti/CompiledMethodLoad/Zombie.java.
>> However, it takes ~3K runs to hit it. I verified the fix by running this
>> test >100K on each platform. Also, I verified that protecting in 'void
>> JvmtiDeferredEventQueue::post(JvmtiEnv* env)' is not enough.
>
> Leonid Mesnik has updated the pull request incrementally with one additional
> commit since the last revision:
>
> post updated.
Looks good!
src/hotspot/share/prims/jvmtiImpl.cpp line 968:
> 966: while (_queue_head != NULL) {
> 967: _queue_head->event().post_compiled_method_load_event(env);
> 968: dequeue();
Good find! So we _can_ zombie the nmethod after we take it off the list.
Makes a lot of sense. Thank you for your perseverance in tracking this down!
-------------
Marked as reviewed by coleenp (Reviewer).
PR: https://git.openjdk.java.net/jdk/pull/4602
