[RFC 3/5] powerpc: atomic: implement atomic{,64}_{add,sub}_return_* variants

[Paul E. McKenney]

On Tue, Sep 01, 2015 at 08:00:27PM +0100, Will Deacon wrote:
> On Fri, Aug 28, 2015 at 04:39:21PM +0100, Peter Zijlstra wrote:
> > On Fri, Aug 28, 2015 at 10:16:02PM +0800, Boqun Feng wrote:
> > > Ah.. just read through the thread you mentioned, I might misunderstand
> > > you, probably because I didn't understand RCpc well..
> > > 
> > > You are saying that in a RELEASE we -might- switch from smp_lwsync() to
> > > smp_mb() semantically, right? I guess this means we -might- switch from
> > > RCpc to RCsc, right?
> > > 
> > > If so, I think I'd better to wait until we have a conclusion for this.
> > 
> > Yes, the difference between RCpc and RCsc is in the meaning of RELEASE +
> > ACQUIRE. With RCsc that implies a full memory barrier, with RCpc it does
> > not.
> 
> We've discussed this before, but for the sake of completeness, I don't
> think we're fully RCsc either because we don't order the actual RELEASE
> operation again a subsequent ACQUIRE operation:
> 
> P0
> smp_store_release(&x, 1);
> foo = smp_load_acquire(&y);
> 
> P1
> smp_store_release(&y, 1);
> bar = smp_load_acquire(&x);
> 
> We allow foo == bar == 0, which is prohibited by SC.

I certainly hope that no one expects foo == bar == 0 to be prohibited!!!

On the other hand, in this case, foo == bar == 1 will be prohibited:

P0
foo = smp_load_acquire(&y);
smp_store_release(&x, 1);

P1
bar = smp_load_acquire(&x);
smp_store_release(&y, 1);

> However, we *do* enforce ordering on any prior or subsequent accesses
> for the code snippet above (the release and acquire combine to give a
> full barrier), which makes these primitives well suited to things like
> message passing.

If I understand your example correctly, neither x86 nor Power implement
a full barrier in this case.  For example:

	P0
	WRITE_ONCE(a, 1);
	smp_store_release(b, 1);
	r1 = smp_load_acquire(c);
	r2 = READ_ONCE(d);

	P1
	WRITE_ONCE(d, 1);
	smp_mb();
	r3 = READ_ONCE(a);

Both x86 and Power can reorder P0 as follows:

	P0
	r1 = smp_load_acquire(c);
	r2 = READ_ONCE(d);
	WRITE_ONCE(a, 1);
	smp_store_release(b, 1);

Which clearly shows that the non-SC outcome r2 == 0 && r3 == 0 is allowed.

Or am I missing your point here?

							Thanx, Paul