[dpdk-dev] [PATCH v2] net/ixgbe: remove MTU setting limitation when port's stopped

[Xing, Beilei]

> -----Original Message-----
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ferruh Yigit
> Sent: Wednesday, July 5, 2017 10:00 PM
> To: Xing, Beilei <beilei.xing at intel.com>; Lu, Wenzhuo
> <wenzhuo.lu at intel.com>
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] net/ixgbe: remove MTU setting
> limitation when port's stopped
> 
> On 7/5/2017 3:33 AM, Beilei Xing wrote:
> > Currently, setting MTU will fail if MTU requires the support of
> > scattered packets before scatter is enabled.
> > This patch allows setting this special MTU when device is stopped,
> > cause scatter will be re-configured during starting port.
> >
> > Signed-off-by: Beilei Xing <beilei.xing at intel.com>
> > ---
> > v2 changes:
> >  - change commit log
> >
> >  drivers/net/ixgbe/ixgbe_ethdev.c | 11 +++++++----
> >  1 file changed, 7 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c
> > b/drivers/net/ixgbe/ixgbe_ethdev.c
> > index fb5574e..ece5d6c 100644
> > --- a/drivers/net/ixgbe/ixgbe_ethdev.c
> > +++ b/drivers/net/ixgbe/ixgbe_ethdev.c
> > @@ -4812,6 +4812,7 @@ ixgbe_dev_mtu_set(struct rte_eth_dev *dev,
> uint16_t mtu)
> >  	struct rte_eth_dev_info dev_info;
> >  	uint32_t frame_size = mtu + ETHER_HDR_LEN + ETHER_CRC_LEN;
> >  	struct rte_eth_rxmode *rx_conf = &dev->data->dev_conf.rxmode;
> > +	struct rte_eth_dev_data *dev_data = dev->data;
> >
> >  	ixgbe_dev_info_get(dev, &dev_info);
> >
> > @@ -4819,13 +4820,15 @@ ixgbe_dev_mtu_set(struct rte_eth_dev *dev,
> uint16_t mtu)
> >  	if ((mtu < ETHER_MIN_MTU) || (frame_size >
> dev_info.max_rx_pktlen))
> >  		return -EINVAL;
> >
> > -	/* refuse mtu that requires the support of scattered packets when
> this
> > -	 * feature has not been enabled before.
> > +	/* If device is started, refuse mtu that requires the support of
> > +	 * scattered packets when this feature has not been enabled before.
> >  	 */
> > -	if (!rx_conf->enable_scatter &&
> > +	if (dev_data->dev_started && !rx_conf->enable_scatter &&
> 
> Should dev->data->scattered_rx be used here?

Agree, I missed this case: if launch testpmd with max_pkt_len bigger than mbuf.
Thanks.

> 
> rx_conf->enable_scatter is configuration value provided by user, but
> dev->data->scattered_rx keeps the actual status of the feature.
> Driver may decide to enable scattered rx even tough configuration didn't
> asked for it, this behavior is what this patch relies on.
> 
> Can you confirm that there is no other case that can prevent enabling
> scattered rx?

I checked that scattered_rx is disabled only when device is stopped.

> Otherwise this patch will let setting MTU bigger than mbuf and if somehow
> scattered rx is not enabled, this may break the Rx path I guess.
> 
> >  	    (frame_size + 2 * IXGBE_VLAN_TAG_SIZE >
> > -	     dev->data->min_rx_buf_size - RTE_PKTMBUF_HEADROOM))
> > +	     dev->data->min_rx_buf_size - RTE_PKTMBUF_HEADROOM)) {
> > +		PMD_INIT_LOG(ERR, "Stop port first.");
> >  		return -EINVAL;
> > +	}
> >
> >  	hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> >  	hlreg0 = IXGBE_READ_REG(hw, IXGBE_HLREG0);
> >