Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Fri, 31 Jan 2020 19:43:22 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     Eric Dumazet <edumazet@google.com>
Cc:     Will Deacon <will@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        "Paul E. McKenney" <paulmck@kernel.org>,
        the arch/x86 maintainers <x86@kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Marco Elver <elver@google.com>
Subject: Re: Confused about hlist_unhashed_lockless()
Message-ID: <20200131184322.GA11457@worktop.programming.kicks-ass.net>
References: <20200131164308.GA5175@willie-the-truck>
 <CANn89i+CnezK81gZSLOy0w7MaZy0uT=xyxoKSTyZU3aMpzifOA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CANn89i+CnezK81gZSLOy0w7MaZy0uT=xyxoKSTyZU3aMpzifOA@mail.gmail.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, Jan 31, 2020 at 08:48:05AM -0800, Eric Dumazet wrote:
>     BUG: KCSAN: data-race in del_timer / detach_if_pending

> diff --git a/include/linux/timer.h b/include/linux/timer.h
> index 1e6650ed066d5d28251b0bd385fc37ef94c96532..0dc19a8c39c9e49a7cde3d34bfa4be8871cbc1c2
> 100644
> --- a/include/linux/timer.h
> +++ b/include/linux/timer.h
> @@ -164,7 +164,7 @@ static inline void destroy_timer_on_stack(struct
> timer_list *timer) { }
>   */
>  static inline int timer_pending(const struct timer_list * timer)
>  {
> - return timer->entry.pprev != NULL;
> + return !hlist_unhashed_lockless(&timer->entry);
>  }

That's just completely wrong.

Aside from any memory barrier issues that might or might not be there
(I'm waaaay to tired atm to tell), the above code is perfectly fine.

In fact, this is a KCSAN compiler infrastructure 'bug'.

Any load that is only compared to zero is immune to load-tearing issues.

The correct thing to do here is something like:

static inline int timer_pending(const struct timer_list *timer)
{
	/*
	 * KCSAN compiler infrastructure is insuffiently clever to
	 * realize that a 'load compared to zero' is immune to
	 * load-tearing.
	 */
	return data_race(timer->entry.pprev != NULL);
}