On Feb 11, 2015 3:15 PM, "Jeremy Fitzhardinge" <jeremy@xxxxxxxx> wrote:
>
> Right now it needs to be a locked operation to prevent read-reordering.
> x86 memory ordering rules state that all writes are seen in a globally
> consistent order, and are globally ordered wrt reads *on the same
> addresses*, but reads to different addresses can be reordered wrt to writes.
The modern x86 rules are actually much tighter than that.
Every store is a release, and every load is an acquire. So a non-atomic store is actually a perfectly fine unlock. All preceding stores will be seen by other cpu's before the unlock, and while reads can pass stores, they only pass *earlier* stores.
For *taking* a lock you need an atomic access, because otherwise loads inside the locked region could bleed out to before the store that takes the lock.
Linus
_______________________________________________ Virtualization mailing list Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linuxfoundation.org/mailman/listinfo/virtualization
