Re: [dpdk-dev] [PATCH] fix ceph async messenger runtime bug based on spdk/dpdk

[Olivier MATZ <olivier.matz@6wind.com>]

Hi Chunmei,

On Wed, Dec 13, 2017 at 03:44:55PM -0800, chunmei Liu wrote:
> From: chunmei <chunmei.liu@intel.com>
> 
> ceph async messenger has some run time error with this dpdk library,
> 1) need set rxm->next= null at end of packet since not init it to null
> when allocate a buffer other wise rte_mbuf_sanity_check will report error.
> 2) when check the size, can't calculate mbuf_data_room_size
> because async messenger dpdk will allocate this part later
> not at create mempool.
> 
> Signed-off-by: chunmei Liu <chunmei.liu@intel.com>
> ---
>  drivers/net/ixgbe/ixgbe_rxtx.c | 3 ++-
>  lib/librte_mbuf/rte_mbuf.c     | 3 ++-
>  2 files changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/ixgbe/ixgbe_rxtx.c b/drivers/net/ixgbe/ixgbe_rxtx.c
> index 1e07895..918946b 100644
> --- a/drivers/net/ixgbe/ixgbe_rxtx.c
> +++ b/drivers/net/ixgbe/ixgbe_rxtx.c
> @@ -2190,7 +2190,8 @@ ixgbe_recv_pkts_lro(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts,
>  			rxm->next = next_rxe->mbuf;
>  			next_sc_entry->fbuf = first_seg;
>  			goto next_desc;
> -		}
> +		} else
> +			rxm->next = NULL;
>  
>  		/* Initialize the first mbuf of the returned packet */
>  		ixgbe_fill_cluster_head_buf(first_seg, &rxd, rxq, staterr);
> diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf.c
> index 0e3e36a..c01da19 100644
> --- a/lib/librte_mbuf/rte_mbuf.c
> +++ b/lib/librte_mbuf/rte_mbuf.c
> @@ -102,7 +102,6 @@ rte_pktmbuf_pool_init(struct rte_mempool *mp, void *opaque_arg)
>  	}
>  
>  	RTE_ASSERT(mp->elt_size >= sizeof(struct rte_mbuf) +
> -		user_mbp_priv->mbuf_data_room_size +
>  		user_mbp_priv->mbuf_priv_size);
>  
>  	mbp_priv = rte_mempool_get_priv(mp);

I don't see the point here.
Many parts of dpdk rely on this code without issue. I enabled the
asserts and see no issue with testpmd. Can you give more details?


> @@ -233,6 +232,8 @@ rte_mbuf_sanity_check(const struct rte_mbuf *m, int is_header)
>  	}
>  	if (nb_segs != 0)
>  		rte_panic("bad nb_segs\n");
> +	if (m_seg != NULL)
> +		rte_panic("bad m_seg\n");
>  }

A similar check was recently sumbitted in another patch. Please see:
https://dpdk.org/dev/patchwork/patch/32057/