> -----Original Message----- > From: Bin Zheng <zhengbin.89...@bytedance.com> > Sent: Friday, December 10, 2021 4:22 PM > To: dev@dpdk.org > Cc: Wang, Haiyue <haiyue.w...@intel.com>; lian...@liangbit.com; > sta...@dpdk.org; Rong, Leyi <leyi.r...@intel.com>; Bin Zheng > <zhengbin.89...@bytedance.com>; jia....@intel.com > Subject: [PATCH v3] net/ixgbe: add vector Rx parameter check > > Under the circumstance that `rx_tail` wrap back to zero and the advance speed > of `rx_tail` is greater than `rxrearm_start`, `rx_tail` will catch up with > `rxrearm_start` and surpass it. > This may cause some mbufs be reused by application. > > So we need to make some restrictions to ensure that `rx_tail` will not exceed > `rxrearm_start`. > > e.g. > > RDH: 972 RDT: 991 rxrearm_nb: 991 rxrearm_start: 992 rx_tail: 959 > RDH: 1004 RDT: 1023 rxrearm_nb: 991 rxrearm_start: 0 rx_tail: 991 > RDH: 12 RDT: 31 rxrearm_nb: 991 rxrearm_start: 32 rx_tail: 1023 > RDH: 31 RDT: 63 rxrearm_nb: 960 rxrearm_start: 64 rx_tail: 0 > RDH: 95 RDT: 95 rxrearm_nb: 1016 rxrearm_start: 96 rx_tail: 88 > RDH: 95 RDT: 127 rxrearm_nb: 991 rxrearm_start: 128 rx_tail: 95 ... > RDH: 908 RDT: 927 rxrearm_nb: 991 rxrearm_start: 928 rx_tail: 895 > RDH: 940 RDT: 959 rxrearm_nb: 991 rxrearm_start: 960 rx_tail: 927 > RDH: 980 RDT: 991 rxrearm_nb: 991 rxrearm_start: 992 rx_tail: 959 > RDH: 991 RDT: 991 rxrearm_nb: 1026 rxrearm_start: 992 rx_tail: 994 > > when `rx_tail` catches up with `rxrearm_start`, > 2(994 - 992) mbufs be reused by application ! > > Bugzilla ID: 882 > Fixes: 5a3cca342417 ("net/ixgbe: fix vector Rx") > Cc: jia....@intel.com > Cc: sta...@dpdk.org > > Signed-off-by: Bin Zheng <zhengbin.89...@bytedance.com> > --- > drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c | 11 +++++++++++ > 1 file changed, 11 insertions(+) > > diff --git a/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c > b/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c > index 1eed949495..4654d0adec 100644 > --- a/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c > +++ b/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c > @@ -364,6 +364,17 @@ _recv_raw_pkts_vec(struct ixgbe_rx_queue *rxq, > struct rte_mbuf **rx_pkts, > uint8_t vlan_flags; > uint16_t udp_p_flag = 0; /* Rx Descriptor UDP header present */ > > + /* > + * Under the circumstance that `rx_tail` wrap back to zero > + * and the advance speed of `rx_tail` is greater than `rxrearm_start`, > + * `rx_tail` will catch up with `rxrearm_start` and surpass it. > + * This may cause some mbufs be reused by application. > + * > + * So we need to make some restrictions to ensure that > + * `rx_tail` will not exceed `rxrearm_start`. > + */ > + nb_pkts = RTE_MIN(nb_pkts, RTE_IXGBE_RXQ_REARM_THRESH); > + > /* nb_pkts has to be floor-aligned to RTE_IXGBE_DESCS_PER_LOOP */ > nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_IXGBE_DESCS_PER_LOOP); > > -- > 2.25.1
Acked-by: Leyi Rong <leyi.r...@intel.com>