[concurrency-interest] Does StampedLock need a releaseFence in theory?

Andrew Haley aph at redhat.com
Mon Jul 25 04:58:09 EDT 2016


On 15/07/16 18:49, Doug Lea wrote:

> Thanks to the ARM folks for finding the tiny bit of wiggle room in
> CAS specs that might in theory (if not in practice) require the
> complementary fence on unlock. Too bad this will slow down some
> implementations for no good reason, but probably not noticeably.

Probably not.  From what I've seen, a StoreStore fence on ARMv8
doesn't cost anything significant when the store buffer is already
empty.  Of course this might change in future designs.

Andrew.


More information about the Concurrency-interest mailing list