RFR: 8206406: StubCodeDesc constructor publishes partially-constructed objects on StubCodeDesc::_list

[Martin Buchholz]

(but memory_order_consume has not been successful.)
http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2016/p0371r0.html


On Fri, Jul 6, 2018 at 5:41 AM, Andrew Haley <aph at redhat.com> wrote:

> On 07/06/2018 03:15 AM, David Holmes wrote:
> > Well ... I never like to see a release_store without a corresponding
> > load_acquire.
>
> Neither do, I, but making _list volatile is enough to ensure that the
> load is effectively a consume operation on every platform supported by
> 8.  We really need to add consume to our memory ordering operators, but
> not for 8.
>
> --
> Andrew Haley
> Java Platform Lead Engineer
> Red Hat UK Ltd. <https://www.redhat.com>
> EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671
>