[dpdk-dev] [PATCH v6 5/9] librte_ether:add data structures of VxLAN filter

[Chilikin, Andrey]

For me these defines make perfect sense - tunnelling filters require 
combinations of different tunnel components, but not all combinations are 
valid. So defining valid combinations separately helps. 

Regards,
Andrey

> -----Original Message-----
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, October 22, 2014 5:31 PM
> To: Liu, Jijiang
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v6 5/9] librte_ether:add data 
> structures of VxLAN filter
> 
> 2014-10-22 02:25, Liu, Jijiang:
> > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > > 2014-10-21 16:46, Jijiang Liu:
> > > > +#define RTE_TUNNEL_FILTER_IMAC_IVLAN (ETH_TUNNEL_FILTER_IMAC
> | \
> > > > +                                       ETH_TUNNEL_FILTER_IVLAN)
> > > > +#define RTE_TUNNEL_FILTER_IMAC_IVLAN_TENID
> (ETH_TUNNEL_FILTER_IMAC | \
> > > > +                                       ETH_TUNNEL_FILTER_IVLAN | \
> > > > +                                       ETH_TUNNEL_FILTER_TENID)
> > > > +#define RTE_TUNNEL_FILTER_IMAC_TENID (ETH_TUNNEL_FILTER_IMAC
> | \
> > > > +                                       ETH_TUNNEL_FILTER_TENID)
> > > > +#define RTE_TUNNEL_FILTER_OMAC_TENID_IMAC
> (ETH_TUNNEL_FILTER_OMAC | \
> > > > +                                       ETH_TUNNEL_FILTER_TENID | \
> > > > +                                       ETH_TUNNEL_FILTER_IMAC)
> > >
> > > I thought you agree that these definitions are useless?
> >
> > Sorry, this MAY be  some misunderstanding, I don't think these 
> > definition are useless. I just thought change "uint16_t filter_type"
> > is better than define "enum filter_type".
> >
> > Let me explain here again.
> > The filter condition are:
> > 1.  inner MAC + inner VLAN
> > 2. inner MAC + IVLAN + tenant ID
> > ..
> > 5. outer MAC + tenant ID + inner MAC
> >
> > For each filter condition, we need to check if the mandatory 
> > parameters are valid by checking corresponding bit MASK.
> 
> Checking bit mask doesn't imply to define all combinations of bit masks.
> There's probably something obvious that one of us is missing.

Anybody else have comments on this? 

> > An pseudo code example:
> >
> >        Switch (filter_type)
> >        Case 1:  //inner MAC + inner VLAN
> >     If (filter_type & ETH_TUNNEL_FILTER_IMAC )
> >                             if   (IMAC==NULL)
> >                                       return -1;
> >
> >        case 5: // outer MAC + tenant ID + inner MAC
> >             If (filter_type & ETH_TUNNEL_FILTER_IMAC )
> >                             if   (IMAC==NULL)
> >                                       return -1;
> >
> >              If (filter_type & ETH_TUNNEL_FILTER_OMAC )
> >                             if   (IMAC==NULL)
> >                                       return -1;
> >    ......