> -----Original Message----- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Yong Liu > Sent: Thursday, July 03, 2014 5:00 PM > To: dev at dpdk.org > Subject: [dpdk-dev] [PATCH] examples/qos_sched: fix flow pause after 2M > packets > > After enable vector pmd, qos_sched only send 32 packets every burst. > That will cause some packets not transmitted and therefore mempool > will be drain after a while. > App qos_sched now will re-send the packets which failed to send out in > previous tx function. > > Signed-off-by: Yong Liu <yong.liu at intel.com> > Acked-by: Cristian Dumitrescu <cristian.dumitrescu at intel.com> > Tested-by: Waterman Cao <waterman.cao at intel.com> > --- > examples/qos_sched/app_thread.c | 16 +++++----------- > 1 files changed, 5 insertions(+), 11 deletions(-) > > diff --git a/examples/qos_sched/app_thread.c > b/examples/qos_sched/app_thread.c > index 7501147..59c4014 100755 > --- a/examples/qos_sched/app_thread.c > +++ b/examples/qos_sched/app_thread.c > @@ -139,17 +139,11 @@ app_send_burst(struct thread_conf *qconf) > > do { > ret = rte_eth_tx_burst(qconf->tx_port, qconf->tx_queue, mbufs, > (uint16_t)n); > - if (unlikely(ret < n)) { /* we cannot drop the packets, so > re-send > */ > - /* update number of packets to be sent */ > - n -= ret; > - mbufs = (struct rte_mbuf **)&mbufs[ret]; > - /* limit number of retries to avoid endless loop */ > - /* reset retry counter if some packets were sent */ > - if (likely(ret != 0)) { > - continue; > - } > - } > - } while (ret != n); > + /* we cannot drop the packets, so re-send */ > + /* update number of packets to be sent */ > + n -= ret; > + mbufs = (struct rte_mbuf **)&mbufs[ret]; > + } while (n); > } > > > -- > 1.7.7.6
How about "mbufs += ret" rather than "mbufs = (struct rte_mbuf **)&mbufs[ret]"?
