From mboxrd@z Thu Jan  1 00:00:00 1970
From: peterz@infradead.org (Peter Zijlstra)
Date: Fri, 30 Sep 2016 21:05:35 +0200
Subject: [RFC] arm64: Enforce observed order for spinlock and data
In-Reply-To: <1475257257-23072-1-git-send-email-bdegraaf@codeaurora.org>
References: <1475257257-23072-1-git-send-email-bdegraaf@codeaurora.org>
Message-ID: <20160930190535.GB3142@twins.programming.kicks-ass.net>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Fri, Sep 30, 2016 at 01:40:57PM -0400, Brent DeGraaf wrote:
> Prior spinlock code solely used load-acquire and store-release
> semantics to ensure ordering of the spinlock lock and the area it
> protects. However, store-release semantics and ordinary stores do
> not protect against accesses to the protected area being observed
> prior to the access that locks the lock itself.
> 
> While the load-acquire and store-release ordering is sufficient
> when the spinlock routines themselves are strictly used, other
> kernel code that references the lock values directly (e.g. lockrefs)

Isn't the problem with lockref the fact that arch_spin_value_unlocked()
isn't a load-acquire, and therefore the CPU in question doesn't need to
observe the contents of the critical section etc..?

That is, wouldn't fixing arch_spin_value_unlocked() by making that an
smp_load_acquire() fix things much better?

> could observe changes to the area protected by the spinlock prior
> to observance of the lock itself being in a locked state, despite
> the fact that the spinlock logic itself is correct.