[PATCH v2 1/5] baseband/acc: fix memory barrier

[Maxime Coquelin]

On 5/2/24 22:45, Hernan Vargas wrote:
> Moving memory barrier so that dequeue thread can be in sync with enqueue
> thread.
> 
> Fixes: 32e8b7ea35dd ("baseband/acc100: refactor to segregate common code")
> Cc: stable at dpdk.org
> 
> Signed-off-by: Hernan Vargas <hernan.vargas at intel.com>
> ---
>   drivers/baseband/acc/acc_common.h | 5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/baseband/acc/acc_common.h b/drivers/baseband/acc/acc_common.h
> index fddeb0737b8b..e249f37e38fe 100644
> --- a/drivers/baseband/acc/acc_common.h
> +++ b/drivers/baseband/acc/acc_common.h
> @@ -1112,6 +1112,9 @@ acc_dma_enqueue(struct acc_queue *q, uint16_t n,
>   				req_elem_addr,
>   				(void *)q->mmio_reg_enqueue);
>   
> +		q->aq_enqueued++;
> +		q->sw_ring_head += enq_batch_size;
> +
>   		rte_wmb();
>   
>   		/* Start time measurement for enqueue function offload. */
> @@ -1122,8 +1125,6 @@ acc_dma_enqueue(struct acc_queue *q, uint16_t n,
>   
>   		queue_stats->acc_offload_cycles += rte_rdtsc_precise() - start_time;
>   
> -		q->aq_enqueued++;
> -		q->sw_ring_head += enq_batch_size;
>   		n -= enq_batch_size;
>   
>   	} while (n);

Reviewed-by: Maxime Coquelin <maxime.coquelin at redhat.com>

Thanks,
Maxime