On Tue, 30 Mar 2021 16:50:50 GMT, Daniel D. Daugherty <dcu...@openjdk.org>
wrote:
>> Robbin Ehn has updated the pull request with a new target base due to a
>> merge or a rebase. The pull request now contains two commits:
>>
>> - Merge branch 'master' into SuspendInHandshake
>> - 8257831: Suspend with handshake (review baseline)
>
> src/hotspot/os/posix/signals_posix.cpp line 1587:
>
>> 1585: // destructor has completed.
>> 1586:
>> 1587: if (thread->is_Java_thread() &&
>> thread->as_Java_thread()->is_terminated()) {
>
> We need @dholmes-ora to verify that this version of the code will
> still solve the bug he was fixing when he added old L1610.
Thanks @dcubed-ojdk - no it won't. The problem is that the signal can hit very
late in a thread's termination process, after the JavaThread destructor has
executed, so no query of JavaThread state is possible. So we needed something
in the Thread state and the SR_lock was a good enough proxy for that. It may be
possible to use a different Thread field (perhaps _ParkEvent but encapsulated
in a Thread::has_terminated() helper method).
-------------
PR: https://git.openjdk.java.net/jdk/pull/3191
