Re: [PATCH v2] tools/memory-model: Add extra ordering for locks and remove it for ordinary release/acquire

From: Will Deacon <will.deacon@arm.com>
To: Alan Stern <stern@rowland.harvard.edu>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Andrea Parri <andrea.parri@amarulasolutions.com>,
	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	LKMM Maintainers -- Akira Yokosawa <akiyks@gmail.com>,
	Boqun Feng <boqun.feng@gmail.com>,
	Daniel Lustig <dlustig@nvidia.com>,
	David Howells <dhowells@redhat.com>,
	Jade Alglave <j.alglave@ucl.ac.uk>,
	Luc Maranget <luc.maranget@inria.fr>,
	Nicholas Piggin <npiggin@gmail.com>,
	Kernel development list <linux-kernel@vger.kernel.org>,
	Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: [PATCH v2] tools/memory-model: Add extra ordering for locks and remove it for ordinary release/acquire
Date: Thu, 12 Jul 2018 18:14:52 +0100	[thread overview]
Message-ID: <20180712171451.GI26935@arm.com> (raw)
In-Reply-To: <Pine.LNX.4.44L0.1807121236470.1306-100000@iolanthe.rowland.org>

Hi Alan,

On Thu, Jul 12, 2018 at 01:04:27PM -0400, Alan Stern wrote:
> On Thu, 12 Jul 2018, Peter Zijlstra wrote:
> > But as you (and Will) point out, we don't so much care about rmw-acquire
> > semantics as much as that we care about unlock+lock behaviour. Another
> > way to look at this is to define:
> > 
> >   smp-store-release + rmw-acquire := TSO (ideally smp_mb)
> > 
> > But then we also have to look at:
> > 
> >   rmw-release + smp-load-acquire
> >   rmw-release + rmw-acquire
> 
> Let's assume that rmw-release is equivalent, in terms of ordering
> strength, to smp_store_release().  Then we can focus our attention on
> just the acquire part.

I can live with that, but it does add another special case, where we could
otherwise just special case acquire/release for the load/store variants
vs everything else.

> On PowerPC, for instance, if spin_lock() used a full HWSYNC fence
> then unlock+lock would become RCsc -- even with no changes to
> spin_unlock().
> 
> > for completeness sake, and I would suggest they result in (at least) the
> > same (TSO) ordering as the one we really care about.
> > 
> > One alternative is to no longer use smp_store_release() for unlock(),
> > and say define atomic_set_release() to be in the rmw-release class
> > instead of being a simple smp_store_release().
> > 
> > Another, and I like this proposal least, is to introduce a new barrier
> > to make this all work.
> 
> This apparently boils down to two questions:
> 
> 	Should spin_lock/spin_unlock be RCsc?

I would love that to be the case, but I'm not asking you to fight that
battle :)

> 	Should rmw-acquire be strong enough so that smp_store_release + 
> 	rmw-acquire is RCtso?
> 
> If both answers are No, we end up with the v3 patch.  If the first
> answer is No and the second is Yes, we end up with the v2 patch.  The
> problem is that different people seem to want differing answers.

Just to be extra unhelpful: I'm happy with either v2 or v3. I suspect Daniel
is the one to convince on v2, because it's RISC-V that's affected by this.

Will