Re: [PATCH v2] membarrier: Document scheduler barrier requirements

[Peter Zijlstra]

On Fri, Aug 18, 2017 at 09:39:16PM -0700, Mathieu Desnoyers wrote:
> @@ -3295,6 +3301,9 @@ static void __sched notrace __schedule(bool preempt)
>  	 * Make sure that signal_pending_state()->signal_pending() below
>  	 * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE)
>  	 * done by the caller to avoid the race with signal_wake_up().
> +	 *
> +	 * The membarrier system call requires a full memory barrier
> +	 * after coming from user-space, before storing to rq->curr.
>  	 */
>  	smp_mb__before_spinlock();
>  	rq_lock(rq, &rf);


Merge conflict here, current tip is over to smp_mb__after_spinlock().