On Thu, 09 Feb 2012 09:36:47 -0800
John Fastabend [off-list ref] wrote:

But the device features makes it easy for user space to learn that the device
supports this sort of offload. Now if all SR-IOV devices support this then it
doesn't matter but I thought there were SR-IOV devices that didn't do any
switching? I'll dig through the SR-IOV drivers to check there are not too
many of them.

If user space needs to know then the OS is not designed properly.
The purpose of the network device is to abstract all those details, and more and more
of them are bleeding through. This makes writing management applications harder and makes
things dependent on features that may or may not be present. The best design is when
the change is invisible.

By netlink_notifier do you mean adding a notifier_block and using atomic_notifier_call_chain()
probably in rtnl_notify()? Then drivers could register with the notifier chain with
atomic_notifier_chain_register() and receive the events correctly. Or did I miss
some notifier chain that already exists?

Yes. that is what I mean. The callbacks you need may or may not already be present.