Re: regression: UFO removal breaks kvm live migration

On Wed, Nov 08, 2017 at 08:32:31PM +0900, David Miller wrote:

From: Jason Wang <redacted>
Date: Wed, 8 Nov 2017 17:25:48 +0900

quoted

On 2017年11月08日 17:08, Willem de Bruijn wrote:

quoted

That won't help in the short term. I'm still reading up to see if
there are
any other options besides reimplement or advertise-but-drop, such as
an implicit trigger that would make the guest renegotiate. It's
unlikely, but
worth a look..

Yes, this looks hard. And even if we can manage to do this, it looks
an overkill since it will impact all guest after migration.

Like Willem I would much prefer "advertise-but-drop" if it works.

In the long term feature renegotiation triggers are a must.

There is no way for us to remove features otherwise.

Isn't this like most userspace ABI issues? Once you add them it's very hard
to remove them, even with negotiation - too much userspace just does
if (ret)
	exit();

In my opinion this will even make migrations more powerful.

I agree. Not sure how to avoid packet drops when doing this though.
And dropping a ton of TX packets isn't nice at all since
TX packets is what teaches the network about the new location
of the VM.

-- 
MST