[dpdk-dev] [PATCH] net/bnx2x: reserve enough headroom for mbuf prepend

[Patil, Harish]

From: zhouyangchao <zhouyates at gmail.com<mailto:zhouyates at gmail.com>>
Date: Wednesday, March 7, 2018 at 10:57 PM
To: Ferruh Yigit <ferruh.yigit at intel.com<mailto:ferruh.yigit at intel.com>>
Cc: "dev at dpdk.org<mailto:dev at dpdk.org>" <dev at dpdk.org<mailto:dev at dpdk.org>>, Harish Patil <Harish.Patil at cavium.com<mailto:Harish.Patil at cavium.com>>, "Mody, Rasesh" <Rasesh.Mody at cavium.com<mailto:Rasesh.Mody at cavium.com>>
Subject: Re: [dpdk-dev] [PATCH] net/bnx2x: reserve enough headroom for mbuf prepend

When allocating a new mbuf for Rx, the value of m->data_off should be
reset to its default value (RTE_PKTMBUF_HEADROOM), instead of reusing
the previous undefined value, which could cause the packet to have a too
small or too high headroom.

On Mon, Mar 5, 2018 at 11:28 PM Ferruh Yigit <ferruh.yigit at intel.com<mailto:ferruh.yigit at intel.com>> wrote:
On 2/6/2018 11:21 AM, zhouyangchao wrote:

Can you please provide more information why this patch is needed?

> Signed-off-by: Yangchao Zhou <zhouyates at gmail.com<mailto:zhouyates at gmail.com>>
> ---
>  drivers/net/bnx2x/bnx2x_rxtx.c | 8 +++++---
>  1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/bnx2x/bnx2x_rxtx.c b/drivers/net/bnx2x/bnx2x_rxtx.c
> index a0d4ac9..d8a3225 100644
> --- a/drivers/net/bnx2x/bnx2x_rxtx.c
> +++ b/drivers/net/bnx2x/bnx2x_rxtx.c
> @@ -140,7 +140,8 @@ bnx2x_dev_rx_queue_setup(struct rte_eth_dev *dev,
>                       return -ENOMEM;
>               }
>               rxq->sw_ring[idx] = mbuf;
> -             rxq->rx_ring[idx] = mbuf->buf_iova;
> +             rxq->rx_ring[idx] =
> +                     rte_cpu_to_le_64(rte_mbuf_data_iova_default(mbuf));
>       }
>       rxq->pkt_first_seg = NULL;
>       rxq->pkt_last_seg = NULL;
> @@ -400,7 +401,8 @@ bnx2x_recv_pkts(void *p_rxq, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
>
>               rx_mb = rxq->sw_ring[bd_cons];
>               rxq->sw_ring[bd_cons] = new_mb;
> -             rxq->rx_ring[bd_prod] = new_mb->buf_iova;
> +             rxq->rx_ring[bd_prod] =
> +                     rte_cpu_to_le_64(rte_mbuf_data_iova_default(new_mb));
>
>               rx_pref = NEXT_RX_BD(bd_cons) & MAX_RX_BD(rxq);
>               rte_prefetch0(rxq->sw_ring[rx_pref]);
> @@ -409,7 +411,7 @@ bnx2x_recv_pkts(void *p_rxq, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
>                       rte_prefetch0(&rxq->sw_ring[rx_pref]);
>               }
>
> -             rx_mb->data_off = pad;
> +             rx_mb->data_off = pad + RTE_PKTMBUF_HEADROOM;
>               rx_mb->nb_segs = 1;
>               rx_mb->next = NULL;
>               rx_mb->pkt_len = rx_mb->data_len = len;
>


Acked-by: Harish Patil <harish.patil at cavium.com<mailto:harish.patil at cavium.com>>