Re: [PATCH net-next v2 2/3] xsk: use atomic operations around cached_prod for copy mode

[Paolo Abeni <pabeni@redhat.com>]

On 11/25/25 9:54 AM, Jason Xing wrote:
> diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
> index 44cc01555c0b..3a023791b273 100644
> --- a/net/xdp/xsk_queue.h
> +++ b/net/xdp/xsk_queue.h
> @@ -402,13 +402,28 @@ static inline void xskq_prod_cancel_n(struct xsk_queue *q, u32 cnt)
>  	q->cached_prod -= cnt;
>  }
>  
> -static inline int xskq_prod_reserve(struct xsk_queue *q)
> +static inline bool xsk_cq_cached_prod_nb_free(struct xsk_queue *q)
>  {
> -	if (xskq_prod_is_full(q))
> +	u32 cached_prod = atomic_read(&q->cached_prod_atomic);
> +	u32 free_entries = q->nentries - (cached_prod - q->cached_cons);
> +
> +	if (free_entries)
> +		return true;
> +
> +	/* Refresh the local tail pointer */
> +	q->cached_cons = READ_ONCE(q->ring->consumer);
> +	free_entries = q->nentries - (cached_prod - q->cached_cons);
> +
> +	return free_entries ? true : false;
> +}
_If_ different CPUs can call xsk_cq_cached_prod_reserve() simultaneously
(as the spinlock existence suggests) the above change introduce a race:

xsk_cq_cached_prod_nb_free() can return true when num_free == 1  on
CPU1, and xsk_cq_cached_prod_reserve increment cached_prod_atomic on
CPU2 before CPU1 completed xsk_cq_cached_prod_reserve().

/P