Re: [Qemu-devel] [PATCH v11 14/15] rdma: introduce MIG_STATE_NONE and change MIG_STATE_SETUP state transition

[Michael R. Hines]

On 06/25/2013 10:55 AM, Paolo Bonzini wrote:
> Il 25/06/2013 16:54, Michael R. Hines ha scritto:
> > > Do those few lines of code change the protocol?  If yes, I'll go against
> > > all my previous recommendations and suggest a #define.  If no, it is
> > > fine to leave it for later, but I would still suggest posting the patch
> > > on the list just for information.
> > Ok, you got it - no it does not change the protocol.
> > I'll use a #define and send it out with a new version for review.
> Make sure you send an additional patch on top of these 15.
>
> Paolo
I was wrong - this does require a protocol extension.

This is because the RDMA transfers are asynchronous, and thus
we cannot know in advance that it is safe to unregister the memory
associated with each individual transfer before the transfer actually
completes.

While the destination currently uses the protocol to participate in
*registering* the page, the destination does not participate in the
RDMA transfers themselves, only the source does, and thus would
require a new exchange of messages to block and instruct the
destination to unpin the memory.

- Michael