On Wed, Aug 23, 2017 at 10:33:57AM +0300, Shahaf Shuler wrote:
> vPMD Tx function assumes that after the scatter of the
> multi-segment packets the next packet will be a single segment packet.
>
> This is not current as the function can return due to lack of resources
> without sending all of the multi-segment mbufs sequence.
>
> Fixes: 6cb559d67b83 ("net/mlx5: add vectorized Rx/Tx burst for x86")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Shahaf Shuler <shah...@mellanox.com>
> ---
>
> This patch should be applied only after the series:
> http://dpdk.org/dev/patchwork/patch/27367/
>
> ---
> drivers/net/mlx5/mlx5_rxtx_vec_sse.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/mlx5/mlx5_rxtx_vec_sse.c
> b/drivers/net/mlx5/mlx5_rxtx_vec_sse.c
> index 8560f745a..30727e6dd 100644
> --- a/drivers/net/mlx5/mlx5_rxtx_vec_sse.c
> +++ b/drivers/net/mlx5/mlx5_rxtx_vec_sse.c
> @@ -119,8 +119,7 @@ txq_wr_dseg_v(struct txq *txq, __m128i *dseg,
> }
>
> /**
> - * Count the number of continuous single segment packets. The first packet
> must
> - * be a single segment packet.
> + * Count the number of continuous single segment packets.
> *
> * @param pkts
> * Pointer to array of packets.
> @@ -137,7 +136,8 @@ txq_check_multiseg(struct rte_mbuf **pkts, uint16_t
> pkts_n)
>
> if (!pkts_n)
> return 0;
> - assert(NB_SEGS(pkts[0]) == 1);
> + if (NB_SEGS(pkts[0]) > 1)
> + return 0;
> /* Count the number of continuous single segment packets. */
> for (pos = 1; pos < pkts_n; ++pos)
> if (NB_SEGS(pkts[pos]) > 1)
> @@ -502,6 +502,8 @@ mlx5_tx_burst_vec(void *dpdk_txq, struct rte_mbuf **pkts,
> uint16_t pkts_n)
> n = RTE_MIN((uint16_t)(pkts_n - nb_tx), MLX5_VPMD_TX_MAX_BURST);
> if (!(txq->flags & ETH_TXQ_FLAGS_NOMULTSEGS))
> n = txq_check_multiseg(&pkts[nb_tx], n);
> + if (!n)
> + break;
> if (!(txq->flags & ETH_TXQ_FLAGS_NOOFFLOADS))
> n = txq_calc_offload(txq, &pkts[nb_tx], n, &cs_flags);
> ret = txq_burst_v(txq, &pkts[nb_tx], n, cs_flags);
> --
> 2.12.0
Acked-by: Nelio Laranjeiro <nelio.laranje...@6wind.com>
--
Nélio Laranjeiro
6WIND
