[ovs-dev] [PATCH v3 2/2] netdev-dpdk: Add option to configure VF MAC address.

[Gaëtan Rivet]

On 28/10/20 02:58 +0100, Gaëtan Rivet wrote:

[...]

> > > 
> > >> +        }
> > >>      }
> > >>      ovs_mutex_unlock(&dev->mutex);
> > >>  
> > >> @@ -1898,6 +1918,7 @@ netdev_dpdk_set_config(struct netdev *netdev, const struct smap *args,
> > >>          {RTE_FC_RX_PAUSE, RTE_FC_FULL    }
> > >>      };
> > >>      const char *new_devargs;
> > >> +    const char *vf_mac;
> > >>      int err = 0;
> > >>  
> > >>      ovs_mutex_lock(&dpdk_mutex);
> > >> @@ -1968,6 +1989,36 @@ netdev_dpdk_set_config(struct netdev *netdev, const struct smap *args,
> > >>          goto out;
> > >>      }
> > >>  
> > >> +    vf_mac = smap_get(args, "dpdk-vf-mac");
> > >> +    if (vf_mac) {
> > >> +        struct eth_addr mac;
> > >> +
> > >> +        err = EINVAL;
> > >> +
> > >> +        if (!dpdk_port_is_representor(dev)) {
> > >> +            VLOG_ERR_BUF(errp, "'%s' is trying to set the VF MAC '%s' "
> > >> +                         "but 'options:dpdk-vf-mac' is only supported for "
> > >> +                         "VF representors.",
> > >> +                         netdev_get_name(netdev), vf_mac);
> > > 
> > > Warnings here seems more in line with the other configs that cannot be
> > > applied.
> > 
> > I agree, if it's not a representor we could just issue a warning and continue.
> > 
> 
> Your phrasing seems to limit the (err -> warn) change to the
> representor filtering part. Should the MAC sanitizing messages below remain
> errors?
> 

I've sent a v4 that followed Kevin's suggestions, let me know if you'd
prefer if I did as I outlined above.

Cheers,
-- 
Gaëtan