Re: [patch] futex: Cure exit race

From: Thomas Gleixner <tglx@linutronix.de>
To: Peter Zijlstra <peterz@infradead.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Stefan Liebler <stli@linux.ibm.com>,
	Heiko Carstens <heiko.carstens@de.ibm.com>,
	Darren Hart <dvhart@infradead.org>,
	Ingo Molnar <mingo@kernel.org>
Subject: Re: [patch] futex: Cure exit race
Date: Tue, 18 Dec 2018 10:31:06 +0100 (CET)	[thread overview]
Message-ID: <alpine.DEB.2.21.1812172239090.1660@nanos.tec.linutronix.de> (raw)
In-Reply-To: <20181212090418.GT5289@hirez.programming.kicks-ass.net>

On Wed, 12 Dec 2018, Peter Zijlstra wrote:
> On Mon, Dec 10, 2018 at 06:43:51PM +0100, Thomas Gleixner wrote:
> @@ -806,6 +806,8 @@ void __noreturn do_exit(long code)
>  		 * task into the wait for ever nirwana as well.
>  		 */
>  		tsk->flags |= PF_EXITPIDONE;
> +		smp_mb();
> +		wake_up_bit(&tsk->flags, 3 /* PF_EXITPIDONE */);

Using ilog2(PF_EXITPIDONE) spares that horrible inline comment and more
importantly selects the right bit. 0x04 is bit 2 ....

> @@ -1187,10 +1236,15 @@ static int attach_to_pi_owner(u32 uval, union futex_key *key,
>  		 * set, we know that the task has finished the
>  		 * cleanup:
>  		 */
>  		int ret = handle_exit_race(uaddr, uval, p);
>  
>  		raw_spin_unlock_irq(&p->pi_lock);
> -		put_task_struct(p);
> +
> +		if (ret == -EAGAIN)
> +			*pe = p;

Hmm, no. We really want to split the return value for that. EAGAIN is also
returned for other reasons.

Plus requeue_pi() needs the same treatment. I'm staring into it, but all I
came up with so far is horribly ugly.

Thanks,

	tglx