Re: [PATCH v2.62] datapath: Add basic MPLS support to kernel

[Simon Horman <horms@verge.net.au>]

On Sat, Jun 28, 2014 at 08:59:01AM -0700, Jesse Gross wrote:
> On Fri, Jun 27, 2014 at 5:55 PM, Simon Horman <horms@verge.net.au> wrote:
> > On Wed, Jun 25, 2014 at 10:51:52AM +0900, Simon Horman wrote:
> >> On Tue, Jun 24, 2014 at 04:24:37PM -0700, Jesse Gross wrote:
> >> >  * Maybe you can refresh my memory - in the case of a push_mpls after
> >> > pop_vlan, why can't we do this check correctly for at least one level
> >> > of vlan? It seems like we could use the inner EtherType to tell
> >> > whether an additional vlan tag is present.
> >>
> >> That seems likely. I'll do some analysis and get back to you.
> >
> > I may be missing something but it seems to me that in
> > ovs_nla_copy_actions__() we have access to the outer EtherType and
> > VLAN TCI but not the inner EtherType.
> >
> > It seems to me that the inner EtherType would need to be present
> > in the key for it to be available in ovs_nla_copy_actions__().
> 
> The inner EtherType (first level) is actually what is stored in
> eth.type in the flow. When eth.tci is set it is assumed that the outer
> EtherType is 0x8100 and that isn't stored.

Yes, of course. Sorry for missing that.