Re: [PATCH 2/3] locking: Clarify requirements for smp_mb__after_spinlock()

From: Peter Zijlstra <peterz@infradead.org>
To: Andrea Parri <andrea.parri@amarulasolutions.com>
Cc: linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org,
	Ingo Molnar <mingo@redhat.com>, Will Deacon <will.deacon@arm.com>,
	Alan Stern <stern@rowland.harvard.edu>,
	Boqun Feng <boqun.feng@gmail.com>,
	Nicholas Piggin <npiggin@gmail.com>,
	David Howells <dhowells@redhat.com>,
	Jade Alglave <j.alglave@ucl.ac.uk>,
	Luc Maranget <luc.maranget@inria.fr>,
	"Paul E . McKenney" <paulmck@linux.vnet.ibm.com>,
	Akira Yokosawa <akiyks@gmail.com>,
	Daniel Lustig <dlustig@nvidia.com>,
	Jonathan Corbet <corbet@lwn.net>,
	Randy Dunlap <rdunlap@infradead.org>
Subject: Re: [PATCH 2/3] locking: Clarify requirements for smp_mb__after_spinlock()
Date: Thu, 28 Jun 2018 17:05:50 +0200	[thread overview]
Message-ID: <20180628150550.GI2494@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <1530182480-13205-3-git-send-email-andrea.parri@amarulasolutions.com>

On Thu, Jun 28, 2018 at 12:41:19PM +0200, Andrea Parri wrote:
> - * This barrier must provide two things:
> - *
> - *   - it must guarantee a STORE before the spin_lock() is ordered against a
> - *     LOAD after it, see the comments at its two usage sites.
> - *
> - *   - it must ensure the critical section is RCsc.
> - *
> - * The latter is important for cases where we observe values written by other
> - * CPUs in spin-loops, without barriers, while being subject to scheduling.
> - *
> - * CPU0			CPU1			CPU2
> - *
> - *			for (;;) {
> - *			  if (READ_ONCE(X))
> - *			    break;
> - *			}
> - * X=1
> - *			<sched-out>
> - *						<sched-in>
> - *						r = X;
> - *
> - * without transitivity it could be that CPU1 observes X!=0 breaks the loop,
> - * we get migrated and CPU2 sees X==0.

Please don't remove that; that explains _why_ we need a full memory
barrier here.

If anything, move it into __schedule() to explain the
smp_mb__after_spinlock() usage there.