Hi, Ferruh Yigit
On 2019/10/15 16:45, Ferruh Yigit wrote: > On 10/9/2019 3:16 PM, Wei Hu (Xavier) wrote: >> From: Hao Chen <chenhao...@huawei.com> >> >> In receiving direction, for FCS error messages, drivers no longer >> record them in rte_eth_stats.ipackets statistics. > If this patch is fixing statistics better to convert commit log into a fix > patch > with Fixes tags etc. This helps for your patches has been backported to LTS. > Same for other patches in the set. > >> In sending direction, for messages of illegal length, too long or >> equals 0, drivers will not notify the network card hardware to >> send them, will not continue to send the remaining message in burst, >> and will not record them in rte_eth_stats.opackets statistics. >> >> Signed-off-by: Hao Chen <chenhao...@huawei.com> >> Signed-off-by: Wei Hu (Xavier) <xavier.hu...@huawei.com> > <...> > >> @@ -979,6 +976,7 @@ hns3_recv_pkts(void *rx_queue, struct rte_mbuf >> **rx_pkts, uint16_t nb_pkts) >> first_seg->pkt_len = pkt_len; >> first_seg->port = rxq->port_id; >> first_seg->hash.rss = rte_le_to_cpu_32(rxdp->rx.rss_hash); >> + first_seg->ol_flags |= PKT_RX_RSS_HASH; > This is unrelated change for this patch, can you please extract into its own > patch? > We will fix them and send patch V2. Thanks for your comment. Regards Xavier
