Re: [PATCH v2 8/9] async: update documentation of the memory barriers

[Richard Henderson]

On 3/6/23 14:33, Paolo Bonzini wrote:
> @@ -107,11 +114,8 @@ static QEMUBH *aio_bh_dequeue(BHList *head, unsigned 
> *flags)
>       QSLIST_REMOVE_HEAD(head, next);
>   
>       /*
> -     * The qatomic_and is paired with aio_bh_enqueue().  The implicit memory
> -     * barrier ensures that the callback sees all writes done by the scheduling
> -     * thread.  It also ensures that the scheduling thread sees the cleared
> -     * flag before bh->cb has run, and thus will call aio_notify again if
> -     * necessary.
> +     * Synchronizes with qatomic_fetch_or() in aio_bh_enqueue(), ensuring that
> +     * the removal finishes before BH_PENDING is reset.
>        */
>       *flags = qatomic_fetch_and(&bh->flags,

Per this new comment, about the remove finishing first, it would seem that we need 
smp_mb__before_rmw here, because QSLIST_REMOVE_HEAD is not SEQCST.


r~