On Wed, Oct 11, 2017 at 01:21:03PM +0100, David Howells wrote: > Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx> wrote: > > > Will Deacon has proposed adding smp_read_barrier_depends() to READ_ONCE(), > > which would mean that quite a few instances of smp_read_barrier_depends() > > would become redundant. > > It's not clear from you description where the barrier is added in relation to > the read: before, after or both? After, similar to lockless_dereference(). Of course, there is not really a barrier except for on Alpha. Thanx, Paul
