On Mon, Nov 21, 2011 at 4:27 PM, Stephen Hemminger <shemmin...@vyatta.com> wrote: > One more comment... > > Shouldn't this device be using netdev_increment_features() like bridging and > bonding > to have the features of the pseudo device reflect those of the underlying > hardware. > This would make the device have TSO only if underlying hardware supported it, > etc.
It probably should in some form. One complication is that Open vSwitch allows multiple internal software devices to be created and people have found various uses for this capability (different vlans and namespaces being a few of the more obvious ones but since the granularity of control is a flow it could represent almost anything). Traffic can flow between these software devices directly and ideally shouldn't be limited by the capabilities of the hardware. Since the current set of offloads is correct, if not always optimal, the thought was that we do this for now and then improve it over time. _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
