RFR: 8252871: fatal error: must own lock JvmtiThreadState_lock

Yasumasa Suenaga ysuenaga at openjdk.java.net
Mon Sep 7 23:09:41 UTC 2020


On Mon, 7 Sep 2020 14:14:28 GMT, Robbin Ehn <rehn at openjdk.org> wrote:

> When these two methods (set_frame_pop/clear_frame_pop) are called in a handshake the requesting thread will have lock
> the JvmtiThreadState_lock. But the thread executing one of these in the handshake may not be the owner.
> So we only check that JvmtiThreadState_lock is locked.
> 
> When verifying the callers to these methods I notice "clear_to_frame_pop" was unused, so instead of fixing it I remove
> it.
> Passes testing locally, still running T3 and T7.
> 
> Now passed!

Thanks for catching up this. Looks good.

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

Marked as reviewed by ysuenaga (Reviewer).

PR: https://git.openjdk.java.net/jdk/pull/60


More information about the serviceability-dev mailing list