> -----Original Message-----
> From: Zhang, Helin
> Sent: Friday, April 6, 2018 11:54 PM
> To: Zhang, Qi Z <qi.z.zh...@intel.com>; eduse...@gmail.com
> Cc: dev@dpdk.org
> Subject: RE: [PATCH] net/i40e/vf: reset scatter_rx flag when configuration
> complies
>
>
>
> > -----Original Message-----
> > From: Zhang, Helin
> > Sent: Friday, April 6, 2018 4:46 PM
> > To: Zhang, Qi Z; eduse...@gmail.com
> > Cc: dev@dpdk.org
> > Subject: RE: [PATCH] net/i40e/vf: reset scatter_rx flag when
> > configuration complies
> >
> >
> >
> > > -----Original Message-----
> > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Zhang, Qi Z
> > > Sent: Monday, April 2, 2018 8:14 PM
> > > To: eduse...@gmail.com
> > > Cc: dev@dpdk.org
> > > Subject: Re: [dpdk-dev] [PATCH] net/i40e/vf: reset scatter_rx flag
> > > when configuration complies
> > >
> > > Hi Serra:
> > >
> > > > -----Original Message-----
> > > > From: eduse...@gmail.com [mailto:eduse...@gmail.com]
> > > > Sent: Friday, March 30, 2018 4:13 AM
> > > > To: Zhang, Qi Z <qi.z.zh...@intel.com>
> > > > Cc: dev@dpdk.org; Eduard Serra <eduse...@gmail.com>
> > > > Subject: [PATCH] net/i40e/vf: reset scatter_rx flag when
> > > > configuration complies
> > > >
> > > > From: Eduard Serra <eduse...@gmail.com>
> > > >
> > > > Scatter RX (scattered_rx) flag is currently not being resetted
> > > > when new configuration is suplied to reconfigure a port, rendering
> > > > the rx callback logic to
> > >
> > > s/suplied/supplied
> > >
> > > > always fall through the scatter branch.
> > > >
> > > > Signed-off-by: Eduard Serra <eduse...@gmail.com>
> > > > ---
> > > > drivers/net/i40e/i40e_ethdev_vf.c | 2 ++
> > > > drivers/net/i40e/i40e_rxtx.c | 2 ++
> > > > 2 files changed, 4 insertions(+)
> > > >
> > > > diff --git a/drivers/net/i40e/i40e_ethdev_vf.c
> > > > b/drivers/net/i40e/i40e_ethdev_vf.c
> > > > index 750d849..a072154 100644
> > > > --- a/drivers/net/i40e/i40e_ethdev_vf.c
> > > > +++ b/drivers/net/i40e/i40e_ethdev_vf.c
> > > > @@ -1750,6 +1750,8 @@ i40evf_rxq_init(struct rte_eth_dev *dev,
> > > > struct i40e_rx_queue *rxq)
> > > > if (dev_data->dev_conf.rxmode.enable_scatter ||
> > > > (rxq->max_pkt_len + 2 * I40E_VLAN_TAG_SIZE) > buf_size) {
> > > > dev_data->scattered_rx = 1;
> > > > + } else {
> > > > + dev_data->scattered_rx = 0;
> > >
> > > Scattered_rx should not be reset if any queue already required You
> > > should reset it before the queue loop
> > Does it mean a NACK?
Yes the fix is not correct.
> >
> > /Helin
> Anyway, please fix the checkpatch warnings reported on patchwork. Thanks!
>
> /Helin
> >
> > >
> > > > }
> > > >
> > > > return 0;
> > > > diff --git a/drivers/net/i40e/i40e_rxtx.c
> > > > b/drivers/net/i40e/i40e_rxtx.c index
> > > > 1217e5a..8698747 100644
> > > > --- a/drivers/net/i40e/i40e_rxtx.c
> > > > +++ b/drivers/net/i40e/i40e_rxtx.c
> > > > @@ -2561,6 +2561,8 @@ i40e_rx_queue_init(struct i40e_rx_queue *rxq)
> > > > /* Check if scattered RX needs to be used. */
> > > > if ((rxq->max_pkt_len + 2 * I40E_VLAN_TAG_SIZE) > buf_size) {
> > > > dev_data->scattered_rx = 1;
> > > > + } else {
> > > > + dev_data->scattered_rx = 0;
> > >
> > > Same as above.
> > >
> > > Regards
> > > Qi
> > >
> > > > }
> > > >
> > > > /* Init the RX tail regieter. */
> > > > --
> > > > 2.7.4
