Re: [Qemu-devel] [PULL] virtio-ccw: dataplane enablement

[Cornelia Huck <cornelia.huck@de.ibm.com>]

On Tue, 16 Jul 2013 15:10:41 +0800
Stefan Hajnoczi <stefanha@gmail.com> wrote:

> On Mon, Jul 15, 2013 at 07:15:55PM +0200, Paolo Bonzini wrote:
> > Il 15/07/2013 18:52, Anthony Liguori ha scritto:
> > > Cornelia Huck <cornelia.huck@de.ibm.com> writes:
> > > 
> > >> The following changes since commit c3cb8e77804313e1be99b5f28a34a346736707a5:
> > >>
> > >>   ioport: remove LITTLE_ENDIAN mark for portio (2013-07-12 14:37:47 -0500)
> > >>
> > >> are available in the git repository at:
> > >>
> > >>   git://github.com/cohuck/qemu virtio-ccw-upstr
> > >>
> > >> for you to fetch changes up to bf72d89f0a8fb27a6bfde1a689690fd577227762:
> > >>
> > >>   virtio-ccw: Enable x-data-plane for virtio-ccw-blk (2013-07-15
> > >>   17:39:04 +0200)
> > > 
> > > This is v1?  This was just posted on Thursday?  That's a bit too quick
> > > for a pull request.
> > > 
> > > I've seen no review from Stefan or Paolo and I would think that given
> > > the work that's been done already, it would be counterproductive to
> > > enable x-data-plane for another backend.
> > > 
> > > Paolo/Stefan, what do you guys think?
> > 
> > Considering what the patch looks like, I don't think it's a huge
> > problem...  In fact, perhaps x-data-plane could be even added to
> > DEFINE_VIRTIO_BLK_PROPERTIES.  This would make it clear that support for
> > non-ioeventfd hosts (including TCG) is one of the things to do to make
> > data plane the default.
> 
> I agree.  Since this patch is so minimal it will not add extra work as
> we reintegrate dataplane into core QEMU.
> 
> Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
> 
Thank you.

Any further guidelines on when to ask for acks from you guys? The mode
I've been operating in is mostly "if virtio-pci provides this feature,
it shouldn't be half bad and we want it in virtio-ccw as well in order
to keep things consistent and get more test coverage".

(dataplane has been enabled in our internal testing for some time now,
I just had asked Dominik to re-do his patch with a proper description
as we had some code churn in our internal repository.)