RFR: 8287788: Implement a better allocator for downcalls [v8]

Maurizio Cimadamore mcimadamore at openjdk.org
Thu Jan 23 15:16:54 UTC 2025


On Thu, 23 Jan 2025 14:54:55 GMT, Maurizio Cimadamore <mcimadamore at openjdk.org> wrote:

>> alternatively - we could have some kind of internal "forceClose" API which doesn't check the owner thread. ~Some care is probably needed as the access is now on a different thread, so we'll need to make sure that some volatile accesses are used by this method, otherwise it might "miss" cleanup actions.~ (no need for volatile as the "carrier" thread is still the same)
>
> A simpler (maybe interim) solution: if the requesting thread is not virtual, use the cache (and use a confined arena). Otherwise use a brand new confined arena.

Yet another option would be to use a confined or shared arena depending on the requesting thread. Which means the terminating thread local would _sometimes_ close the arena, sometimes it will leave it to the GC.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/23142#discussion_r1927150864


More information about the core-libs-dev mailing list