Linux Virtualization - Mar 2023 - [PATCH v2] virtio: add VIRTIO_F_NOTIFICATION_DATA feature support

On Tue, Mar 21, 2023 at 12:00:42PM +0300, Viktor Prutyanov
wrote:
> On Tue, Mar 21, 2023 at 5:29?AM Jason Wang <jasowang at redhat.com>
wrote:
> >
> > On Tue, Mar 21, 2023 at 7:21?AM Viktor Prutyanov <viktor at
daynix.com> wrote:
> > >
> > > According to VirtIO spec v1.2, VIRTIO_F_NOTIFICATION_DATA feature
> > > indicates that the driver passes extra data along with the queue
> > > notifications.
> > >
> > > In a split queue case, the extra data is 16-bit available index.
In a
> > > packed queue case, the extra data is 1-bit wrap counter and
15-bit
> > > available index.
> > >
> > > Add support for this feature for MMIO and PCI transports. Channel
I/O
> > > transport will not accept this feature.
> > >
> > > Signed-off-by: Viktor Prutyanov <viktor at daynix.com>
> > > ---
> > >
> > >  v2: reject the feature in virtio_ccw, replace __le32 with u32
> > >
> > >  drivers/s390/virtio/virtio_ccw.c   |  4 +---
> > >  drivers/virtio/virtio_mmio.c       | 15 ++++++++++++++-
> > >  drivers/virtio/virtio_pci_common.c | 10 ++++++++++
> > >  drivers/virtio/virtio_pci_common.h |  4 ++++
> > >  drivers/virtio/virtio_pci_legacy.c |  2 +-
> > >  drivers/virtio/virtio_pci_modern.c |  2 +-
> > >  drivers/virtio/virtio_ring.c       | 17 +++++++++++++++++
> > >  include/linux/virtio_ring.h        |  2 ++
> > >  include/uapi/linux/virtio_config.h |  6 ++++++
> > >  9 files changed, 56 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/drivers/s390/virtio/virtio_ccw.c
b/drivers/s390/virtio/virtio_ccw.c
> > > index a10dbe632ef9..d72a59415527 100644
> > > --- a/drivers/s390/virtio/virtio_ccw.c
> > > +++ b/drivers/s390/virtio/virtio_ccw.c
> > > @@ -789,9 +789,7 @@ static u64 virtio_ccw_get_features(struct
virtio_device *vdev)
> > >
> > >  static void ccw_transport_features(struct virtio_device *vdev)
> > >  {
> > > -       /*
> > > -        * Currently nothing to do here.
> > > -        */
> > > +       __virtio_clear_bit(vdev, VIRTIO_F_NOTIFICATION_DATA);
> >
> > Is there any restriction that prevents us from implementing
> > VIRTIO_F_NOTIFICATION_DATA? (Spec seems doesn't limit us from
this)
> 
> Most likely, nothing.
So pls code it up. It's the same format.
> >
> > >  }
> > >
> > >  static int virtio_ccw_finalize_features(struct virtio_device
*vdev)
> > > diff --git a/drivers/virtio/virtio_mmio.c
b/drivers/virtio/virtio_mmio.c
> > > index 3ff746e3f24a..0e13da17fe0a 100644
> > > --- a/drivers/virtio/virtio_mmio.c
> > > +++ b/drivers/virtio/virtio_mmio.c
> > > @@ -285,6 +285,19 @@ static bool vm_notify(struct virtqueue *vq)
> > >         return true;
> > >  }
> > >
> > > +static bool vm_notify_with_data(struct virtqueue *vq)
> > > +{
> > > +       struct virtio_mmio_device *vm_dev =
to_virtio_mmio_device(vq->vdev);
> > > +       u32 data = vring_fill_notification_data(vq);
> >
> > Can we move this to the initialization?
> 
> This data is new for each notification, because it helps to identify
> the next available index.
> 
> >
> > Thanks
> >
> 
> Thanks,
> Viktor Prutyanov

On Tue, Mar 21 2023, "Michael S. Tsirkin" <mst at redhat.com>
wrote:
> On Tue, Mar 21, 2023 at 12:00:42PM +0300, Viktor Prutyanov wrote:
>> On Tue, Mar 21, 2023 at 5:29?AM Jason Wang <jasowang at
redhat.com> wrote:
>> >
>> > On Tue, Mar 21, 2023 at 7:21?AM Viktor Prutyanov <viktor at
daynix.com> wrote:
>> > >
>> > > According to VirtIO spec v1.2, VIRTIO_F_NOTIFICATION_DATA
feature
>> > > indicates that the driver passes extra data along with the
queue
>> > > notifications.
>> > >
>> > > In a split queue case, the extra data is 16-bit available
index. In a
>> > > packed queue case, the extra data is 1-bit wrap counter and
15-bit
>> > > available index.
>> > >
>> > > Add support for this feature for MMIO and PCI transports.
Channel I/O
>> > > transport will not accept this feature.
>> > >
>> > > Signed-off-by: Viktor Prutyanov <viktor at daynix.com>
>> > > ---
>> > >
>> > >  v2: reject the feature in virtio_ccw, replace __le32 with
u32
>> > >
>> > >  drivers/s390/virtio/virtio_ccw.c   |  4 +---
>> > >  drivers/virtio/virtio_mmio.c       | 15 ++++++++++++++-
>> > >  drivers/virtio/virtio_pci_common.c | 10 ++++++++++
>> > >  drivers/virtio/virtio_pci_common.h |  4 ++++
>> > >  drivers/virtio/virtio_pci_legacy.c |  2 +-
>> > >  drivers/virtio/virtio_pci_modern.c |  2 +-
>> > >  drivers/virtio/virtio_ring.c       | 17 +++++++++++++++++
>> > >  include/linux/virtio_ring.h        |  2 ++
>> > >  include/uapi/linux/virtio_config.h |  6 ++++++
>> > >  9 files changed, 56 insertions(+), 6 deletions(-)
>> > >
>> > > diff --git a/drivers/s390/virtio/virtio_ccw.c
b/drivers/s390/virtio/virtio_ccw.c
>> > > index a10dbe632ef9..d72a59415527 100644
>> > > --- a/drivers/s390/virtio/virtio_ccw.c
>> > > +++ b/drivers/s390/virtio/virtio_ccw.c
>> > > @@ -789,9 +789,7 @@ static u64 virtio_ccw_get_features(struct
virtio_device *vdev)
>> > >
>> > >  static void ccw_transport_features(struct virtio_device
*vdev)
>> > >  {
>> > > -       /*
>> > > -        * Currently nothing to do here.
>> > > -        */
>> > > +       __virtio_clear_bit(vdev, VIRTIO_F_NOTIFICATION_DATA);
>> >
>> > Is there any restriction that prevents us from implementing
>> > VIRTIO_F_NOTIFICATION_DATA? (Spec seems doesn't limit us from
this)
>> 
>> Most likely, nothing.
>
> So pls code it up. It's the same format.
FWIW, the notification data needs to go via the third parameter of
kvm_hypercall3() in virtio_ccw_kvm_notify().